2020.07.20.txt

==========New Papers==========
1, TITLE:       SiamParseNet: Joint Body Parsing and Label Propagation in Infant Movement Videos
http://arxiv.org/abs/2007.08646
AUTHORS:        Haomiao Ni ; Yuan Xue ; Qian Zhang ; Xiaolei Huang
COMMENTS:       MICCAI 2020
HIGHLIGHT:      In this paper, we propose a semi-supervised body parsing model, termed SiamParseNet (SPN), to jointly learn single frame body parsing and label propagation between frames in a semi-supervised fashion.

2, TITLE:       FSpiNN: An Optimization Framework for Memory- and Energy-Efficient Spiking Neural Networks
http://arxiv.org/abs/2007.08860
AUTHORS:        Rachmad Vidya Wicaksana Putra ; Muhammad Shafique
COMMENTS:       To appear at the IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (IEEE-TCAD), as part of the ESWEEK-TCAD Special Issue, September 2020
HIGHLIGHT:      Towards this, we propose FSpiNN, an optimization framework for obtaining memory- and energy-efficient SNNs for training and inference processing, with unsupervised learning capability while maintaining accuracy.

3, TITLE:       Preserving Semantic Neighborhoods for Robust Cross-modal Retrieval
http://arxiv.org/abs/2007.08617
AUTHORS:        Christopher Thomas ; Adriana Kovashka
HIGHLIGHT:      We propose novel within-modality losses which encourage semantic coherency in both the text and image subspaces, which does not necessarily align with visual coherency.

4, TITLE:       EPNet: Enhancing Point Features with Image Semantics for 3D Object Detection
http://arxiv.org/abs/2007.08856
AUTHORS:        Tengteng Huang ; Zhe Liu ; Xiwu Chen ; Xiang Bai
COMMENTS:       Accepted by ECCV2020
HIGHLIGHT:      In this paper, we aim at addressing two critical issues in the 3D detection task, including the exploitation of multiple sensors~(namely LiDAR point cloud and camera image), as well as the inconsistency between the localization and classification confidence.

5, TITLE:       Dynamic Low-light Imaging with Quanta Image Sensors
http://arxiv.org/abs/2007.08614
AUTHORS:        Yiheng Chi ; Abhiram Gnanasambandam ; Vladlen Koltun ; Stanley H. Chan
COMMENTS:       Published in the 16th European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      We propose a solution using Quanta Image Sensors (QIS) and present a new image reconstruction algorithm.

6, TITLE:       A biological plausible audio-visual integration model for continual lifelong learning
http://arxiv.org/abs/2007.08855
AUTHORS:        Wenjie Chen ; Fengtong Du ; Ye Wang ; Lihong Cao
HIGHLIGHT:      Here we assume that the integration of audio and visual perceptual information in the MTL during learning is a crucial step to form concepts and make continual learning possible, and we propose a biological plausible audio-visual integration model (AVIM), which is a spiking neural network with multi-compartmental neuron model and a calcium based synaptic tagging and capture plasticity model, as a possible mechanism of concept formation.

7, TITLE:       Collision Avoidance Robotics Via Meta-Learning (CARML)
http://arxiv.org/abs/2007.08616
AUTHORS:        Abhiram Iyer ; Aravind Mahadevan
HIGHLIGHT:      This paper presents an approach to exploring a multi-objective reinforcement learning problem with Model-Agnostic Meta-Learning.

8, TITLE:       Explainable Deep Learning for Uncovering Actionable Scientific Insights for Materials Discovery and Design
http://arxiv.org/abs/2007.08631
AUTHORS:        Shusen Liu ; Bhavya Kailkhura ; Jize Zhang ; Anna M. Hiszpanski ; Emily Robertson ; Donald Loveland ; T. Yong-Jin Han
HIGHLIGHT:      In this work, we propose techniques for exploring the behavior of deep learning models by injecting domain-specific actionable attributes as tunable "knobs" in the analysis pipeline.

9, TITLE:       Impact of base dataset design on few-shot image classification
http://arxiv.org/abs/2007.08872
AUTHORS:        Othman Sbai ; Camille Couprie ; Mathieu Aubry
COMMENTS:       23 pages, 11 figures, to appear in ECCV 2020
HIGHLIGHT:      In this paper, we systematically study the effect of variations in the training data by evaluating deep features trained on different image sets in a few-shot classification setting.

10, TITLE:      Scale Equivariance Improves Siamese Tracking
http://arxiv.org/abs/2007.09115
AUTHORS:        Ivan Sosnovik ; Artem Moskalev ; Arnold Smeulders
HIGHLIGHT:      In this paper, we focus on scaling and we aim to equip the Siamese network with additional built-in scale equivariance to capture the natural variations of the target a priori.

11, TITLE:      DVI: Depth Guided Video Inpainting for Autonomous Driving
http://arxiv.org/abs/2007.08854
AUTHORS:        Miao Liao ; Feixiang Lu ; Dingfu Zhou ; Sibo Zhang ; Wei Li ; Ruigang Yang
HIGHLIGHT:      To get clear street-view and photo-realistic simulation in autonomous driving, we present an automatic video inpainting algorithm that can remove traffic agents from videos and synthesize missing regions with the guidance of depth/point cloud. To verify the effectiveness of our approach, we build a large inpainting dataset in the real urban road environment with synchronized images and Lidar data including many challenge scenes, e.g., long time occlusion.

12, TITLE:      Understanding and Diagnosing Vulnerability under Adversarial Attacks
http://arxiv.org/abs/2007.08716
AUTHORS:        Haizhong Zheng ; Ziqi Zhang ; Honglak Lee ; Atul Prakash
HIGHLIGHT:      In this work, we propose a novel interpretability method, InterpretGAN, to generate explanations for features used for classification in latent variables.

13, TITLE:      Constraint-Based Software Diversification for Efficient Mitigation of Code-Reuse Attacks
http://arxiv.org/abs/2007.08955
AUTHORS:        Rodothea Myrsini Tsoupidi ; Roberto Castañeda Lozano ; Benoit Baudry
COMMENTS:       15 pages, 26th International Conference on Principles and Practice of Constraint Programming
HIGHLIGHT:      This paper introduces Diversity by Construction (DivCon), a constraint-based compiler approach to software diversification.

14, TITLE:      2nd Place Solution to ECCV 2020 VIPriors Object Detection Challenge
http://arxiv.org/abs/2007.08849
AUTHORS:        Yinzheng Gu ; Yihan Pan ; Shizhe Chen
COMMENTS:       Technical report for the ECCV 2020 VIPriors Object Detection Challenge
HIGHLIGHT:      In this report, we descibe our approach to the ECCV 2020 VIPriors Object Detection Challenge which took place from March to July in 2020.

15, TITLE:      SummPip: Unsupervised Multi-Document Summarization with Sentence Graph Compression
http://arxiv.org/abs/2007.08954
AUTHORS:        Jinming Zhao ; Ming Liu ; Longxiang Gao ; Yuan Jin ; Lan Du ; He Zhao ; He Zhang ; Gholamreza Haffari
COMMENTS:       accepted to SIGIR 2020
HIGHLIGHT:      Obtaining training data for multi-document summarization (MDS) is time consuming and resource-intensive, so recent neural models can only be trained for limited domains.

16, TITLE:      CovidCare: Transferring Knowledge from Existing EMR to Emerging Epidemic for Interpretable Prognosis
http://arxiv.org/abs/2007.08848
AUTHORS:        Liantao Ma ; Xinyu Ma ; Junyi Gao ; Chaohe Zhang ; Zhihao Yu ; Xianfeng Jiao ; Wenjie Ruan ; Yasha Wang ; Wen Tang ; Jiangtao Wang
HIGHLIGHT:      In this paper, we propose a deep-learning-based approach, CovidCare, which leverages the existing electronic medical records to enhance the prognosis for inpatients with emerging infectious diseases.

17, TITLE:      Transfer Learning without Knowing: Reprogramming Black-box Machine Learning Models with Scarce Data and Limited Resources
http://arxiv.org/abs/2007.08714
AUTHORS:        Yun-Yun Tsai ; Pin-Yu Chen ; Tsung-Yi Ho
COMMENTS:       Totally 13 pages, including 9 pages for main paper, 2 extra pages for references, and 2 extra pages for supplementary material
HIGHLIGHT:      Motivated by the techniques from adversarial machine learning (ML) that are capable of manipulating the model prediction via data perturbations, in this paper we propose a novel approach, black-box adversarial reprogramming (BAR), that repurposes a well-trained black-box ML model (e.g., a prediction API or a proprietary software) for solving different ML tasks, especially in the scenario with scarce data and constrained resources.

18, TITLE:      AlignNet: Unsupervised Entity Alignment
http://arxiv.org/abs/2007.08973
AUTHORS:        Antonia Creswell ; Kyriacos Nikiforou ; Oriol Vinyals ; Andre Saraiva ; Rishabh Kabra ; Loic Matthey ; Chris Burgess ; Malcolm Reynolds ; Richard Tanburn ; Marta Garnelo ; Murray Shanahan
HIGHLIGHT:      In this paper we take steps towards solving the alignment problem, presenting the AlignNet, an unsupervised alignment module.

19, TITLE:      LEED: Label-Free Expression Editing via Disentanglement
http://arxiv.org/abs/2007.08971
AUTHORS:        Rongliang Wu ; Shijian Lu
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      This paper presents an innovative label-free expression editing via disentanglement (LEED) framework that is capable of editing the expression of both frontal and profile facial images without requiring any expression label.

20, TITLE:      Synthetic and Real Inputs for ToolSegmentation in Robotic Surgery
http://arxiv.org/abs/2007.09107
AUTHORS:        Emanuele Colleoni ; Philip Edwards ; Danail Stoyanov
HIGHLIGHT:      At present labelled data for training deep learning models is still lacking for semantic surgical instrument segmentation and in this paper we show that it may be possible to use robot kinematic data coupled with laparoscopic images to alleviate the labelling problem.

21, TITLE:      Compositional Generalization in Semantic Parsing: Pre-training vs. Specialized Architectures
http://arxiv.org/abs/2007.08970
AUTHORS:        Daniel Furrer ; Marc van Zee ; Nathan Scales ; Nathanael Schärli
HIGHLIGHT:      We show that masked language model (MLM) pre-training rivals SCAN-inspired architectures on primitive holdout splits.

22, TITLE:      Detecting Human-Object Interactions with Action Co-occurrence Priors
http://arxiv.org/abs/2007.08728
AUTHORS:        Dong-Jin Kim ; Xiao Sun ; Jinsoo Choi ; Stephen Lin ; In So Kweon
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper, we model the correlations as action co-occurrence matrices and present techniques to learn these priors and leverage them for more effective training, especially in rare classes.

23, TITLE:      Weakly-supervised Learning of Human Dynamics
http://arxiv.org/abs/2007.08969
AUTHORS:        Petrissa Zell ; Bodo Rosenhahn ; Bastian Wandt
HIGHLIGHT:      This paper proposes a weakly-supervised learning framework for dynamics estimation from human motion.

24, TITLE:      End-to-end Deep Prototype and Exemplar Models for Predicting Human Behavior
http://arxiv.org/abs/2007.08723
AUTHORS:        Pulkit Singh ; Joshua C. Peterson ; Ruairidh M. Battleday ; Thomas L. Griffiths
COMMENTS:       7 pages, 4 figures, 2 tables. Accepted as a paper to the 42nd Annual Meeting of the Cognitive Science Society (CogSci 2020)
HIGHLIGHT:      In this work, we extend classic prototype and exemplar models to learn both stimulus and category representations jointly from raw input.

25, TITLE:      On Robustness and Transferability of Convolutional Neural Networks
http://arxiv.org/abs/2007.08558
AUTHORS:        Josip Djolonga ; Jessica Yung ; Michael Tschannen ; Rob Romijnders ; Lucas Beyer ; Alexander Kolesnikov ; Joan Puigcerver ; Matthias Minderer ; Alexander D'Amour ; Dan Moldovan ; Sylvan Gelly ; Neil Houlsby ; Xiaohua Zhai ; Mario Lucic
HIGHLIGHT:      In this work we revisit the out-of-distribution and transfer performance of modern image classification CNNs and investigate the impact of the pre-training data size, the model scale, and the data preprocessing pipeline. Finally, we outline the shortcomings of existing robustness evaluation datasets and introduce a synthetic dataset we use for a systematic analysis across common factors of variation.

26, TITLE:      Re-weighting and 1-Point RANSAC-Based PnP Solution to Handle Outliers
http://arxiv.org/abs/2007.08577
AUTHORS:        Haoyin Zhou ; Tao Zhang ; Jagadeesan Jayender
COMMENTS:       https://github.com/haoyinzhou/PnP_Toolbox
HIGHLIGHT:      We propose a fast PnP solution named R1PPnP to handle outliers by utilizing a soft re-weighting mechanism and the 1-point RANSAC scheme.

27, TITLE:      Geometric Correspondence Fields: Learned Differentiable Rendering for 3D Pose Refinement in the Wild
http://arxiv.org/abs/2007.08939
AUTHORS:        Alexander Grabner ; Yaming Wang ; Peizhao Zhang ; Peihong Guo ; Tong Xiao ; Peter Vajda ; Peter M. Roth ; Vincent Lepetit
COMMENTS:       Accepted to European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      We present a novel 3D pose refinement approach based on differentiable rendering for objects of arbitrary categories in the wild.

28, TITLE:      Real-time Surface Deformation Recovery from Stereo Videos
http://arxiv.org/abs/2007.08576
AUTHORS:        Haoyin Zhou ; Jagadeesan Jayender
COMMENTS:       In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) (pp. 339-347). Springer, Cham
HIGHLIGHT:      In this paper, we propose an approach to estimate the deformation of tissue surface from stereo videos in real-time, which is capable of handling occlusion, smooth surface and fast deformation.

29, TITLE:      Camera Bias in a Fine Grained Classification Task
http://arxiv.org/abs/2007.08574
AUTHORS:        Philip T. Jackson ; Stephen Bonner ; Ning Jia ; Christopher Holder ; Jon Stonehouse ; Boguslaw Obara
HIGHLIGHT:      We show that correlations between the camera used to acquire an image and the class label of that image can be exploited by convolutional neural networks (CNN), resulting in a model that "cheats" at an image classification task by recognizing which camera took the image and inferring the class label from the camera.

30, TITLE:      Visualizing the Finer Cluster Structure of Large-Scale and High-Dimensional Data
http://arxiv.org/abs/2007.08711
AUTHORS:        Yu Liang ; Arin Chaudhuri ; Haoyu Wang
HIGHLIGHT:      In this paper, we propose using a generalized sigmoid function to model the distance similarity in both high- and low-dimensional spaces.

31, TITLE:      Cross-Identity Motion Transfer for Arbitrary Objects through Pose-Attentive Video Reassembling
http://arxiv.org/abs/2007.08786
AUTHORS:        Subin Jeon ; Seonghyeon Nam ; Seoung Wug Oh ; Seon Joo Kim
COMMENTS:       ECCV 2020
HIGHLIGHT:      We propose an attention-based networks for transferring motions between arbitrary objects.

32, TITLE:      Learning Posterior and Prior for Uncertainty Modeling in Person Re-Identification
http://arxiv.org/abs/2007.08785
AUTHORS:        Yan Zhang ; Zhilin Zheng ; Binyu He ; Li Sun
HIGHLIGHT:      This paper proposes to learn the sample posterior and the class prior distribution in the latent space, so that not only representative features but also the uncertainty can be built by the model.

33, TITLE:      Mixing Real and Synthetic Data to Enhance Neural Network Training -- A Review of Current Approaches
http://arxiv.org/abs/2007.08781
AUTHORS:        Viktor Seib ; Benjamin Lange ; Stefan Wirtz
HIGHLIGHT:      In this work we review and compare different techniques available in the literature to improve training results without acquiring additional annotated real-world data.

34, TITLE:      DACS: Domain Adaptation via Cross-domain Mixed Sampling
http://arxiv.org/abs/2007.08702
AUTHORS:        Wilhelm Tranheden ; Viktor Olsson ; Juliano Pinto ; Lennart Svensson
HIGHLIGHT:      We propose DACS: Domain Adaptation via Cross-domain mixed Sampling, which mixes images from the two domains along with the corresponding labels.

35, TITLE:      HDNet: Human Depth Estimation for Multi-Person Camera-Space Localization
http://arxiv.org/abs/2007.08943
AUTHORS:        Jiahao Lin ; Gim Hee Lee
COMMENTS:       16 pages, 5 figures. Accepted in ECCV 2020
HIGHLIGHT:      In this paper, we propose the Human Depth Estimation Network (HDNet), an end-to-end framework for absolute root joint localization in the camera coordinate space.

36, TITLE:      Deep Learning Based Traffic Surveillance System For Missing and Suspicious Car Detection
http://arxiv.org/abs/2007.08783
AUTHORS:        K. V. Kadambari ; Vishnu Vardhan Nimmalapudi
HIGHLIGHT:      This paper presents a deep learning based automatic traffic surveillance system for the detection of stolen/suspicious cars from the closed circuit television(CCTV) camera footage.

37, TITLE:      Vision-based Estimation of MDS-UPDRS Gait Scores for Assessing Parkinson's Disease Motor Severity
http://arxiv.org/abs/2007.08920
AUTHORS:        Mandy Lu ; Kathleen Poston ; Adolf Pfefferbaum ; Edith V. Sullivan ; Li Fei-Fei ; Kilian M. Pohl ; Juan Carlos Niebles ; Ehsan Adeli
COMMENTS:       Accepted as a conference paper at MICCAI (Medical Image Computing and Computer Assisted Intervention), Lima, Peru, October 2020. 11 pages, LaTeX
HIGHLIGHT:      For the first time, we propose a computer vision-based model that observes non-intrusive video recordings of individuals, extracts their 3D body skeletons, tracks them through time, and classifies the movements according to the MDS-UPDRS gait scores.

38, TITLE:      Grad-Cam Guided Progressive Feature CutMix for Classification
http://arxiv.org/abs/2007.08779
AUTHORS:        Yan Zhang ; Binyu He ; Li Sun
HIGHLIGHT:      This paper deals with this issue by performing the attentive feature cutmix in a progressive manner, among the multi-branch classifier trained on the same task.

39, TITLE:      Edge-Preserving Guided Semantic Segmentation for VIPriors Challenge
http://arxiv.org/abs/2007.08919
AUTHORS:        Chih-Chung Hsu ; Hsin-Ti Ma
COMMENTS:       Technical report for VIPChallenge
HIGHLIGHT:      To overcome this shortcoming, therefore, we propose edge-preserving guidance to obtain the extra prior information, to avoid the overfitting under small-scale training dataset.

40, TITLE:      A New Look at Ghost Normalization
http://arxiv.org/abs/2007.08554
AUTHORS:        Neofytos Dimitriou ; Ognjen Arandjelovic
HIGHLIGHT:      Our contributions are: (i) we uncover a source of regularization that is unique to GhostNorm, and not simply an extension from BatchNorm, (ii) three types of GhostNorm implementations are described, two of which employ BatchNorm as the underlying normalization technique, (iii) by visualising the loss landscape of GhostNorm, we observe that GhostNorm consistently decreases the smoothness when compared to BatchNorm, (iv) we introduce Sequential Normalization (SeqNorm), and report superior performance over state-of-the-art methodologies on both CIFAR--10 and CIFAR--100 datasets.

41, TITLE:      InfoFocus: 3D Object Detection for Autonomous Driving with Dynamic Information Modeling
http://arxiv.org/abs/2007.08556
AUTHORS:        Jun Wang ; Shiyi Lan ; Mingfei Gao ; Larry S. Davis
HIGHLIGHT:      To address this issue, we propose a novel 3D object detection framework with dynamic information modeling.

42, TITLE:      Smooth Deformation Field-based Mismatch Removal in Real-time
http://arxiv.org/abs/2007.08553
AUTHORS:        Haoyin Zhou ; Jagadeesan Jayender
COMMENTS:       submitted for peer review since 10/2019
HIGHLIGHT:      This paper studies the mismatch removal problem, which may serve as the subsequent step of feature matching.

43, TITLE:      Discovering Reinforcement Learning Algorithms
http://arxiv.org/abs/2007.08794
AUTHORS:        Junhyuk Oh ; Matteo Hessel ; Wojciech M. Czarnecki ; Zhongwen Xu ; Hado van Hasselt ; Satinder Singh ; David Silver
HIGHLIGHT:      This paper introduces a new meta-learning approach that discovers an entire update rule which includes both 'what to predict' (e.g. value functions) and 'how to learn from it' (e.g. bootstrapping) by interacting with a set of environments.

44, TITLE:      Explanation-Guided Training for Cross-Domain Few-Shot Classification
http://arxiv.org/abs/2007.08790
AUTHORS:        Jiamei Sun ; Sebastian Lapuschkin ; Wojciech Samek ; Yunqing Zhao ; Ngai-Man Cheung ; Alexander Binder
HIGHLIGHT:      In this paper, we introduce a novel training approach for existing FSC models.

45, TITLE:      Multi-Classifier selection-fusion framework: application to NDT of complex metallic parts
http://arxiv.org/abs/2007.08789
AUTHORS:        Vahid Yaghoubi ; Liangliang Cheng ; Wim Van Paepegem ; Mathias Kersemans
HIGHLIGHT:      In this paper, a multi classifier selection-fusion framework based on the Dempster-Shafer theory is proposed.

46, TITLE:      Talking-head Generation with Rhythmic Head Motion
http://arxiv.org/abs/2007.08547
AUTHORS:        Lele Chen ; Guofeng Cui ; Celong Liu ; Zhong Li ; Ziyi Kou ; Yi Xu ; Chenliang Xu
HIGHLIGHT:      To overcome the limitations, we propose a 3D-aware generative network along with a hybrid embedding module and a non-linear composition module.

47, TITLE:      AE-Net: Autonomous Evolution Image Fusion Method Inspired by Human Cognitive Mechanism
http://arxiv.org/abs/2007.08763
AUTHORS:        Aiqing Fang ; Xinbo Zhao ; Jiaqi Yang ; Shihao Cao ; Yanning Zhang
HIGHLIGHT:      In order to solve the robustness and generality problems of the image fusion task,inspired by the human brain cognitive mechanism, we propose a robust and general image fusion method with autonomous evolution ability, and is therefore denoted with AE-Net.

48, TITLE:      Sketching Image Gist: Human-Mimetic Hierarchical Scene Graph Generation
http://arxiv.org/abs/2007.08760
AUTHORS:        Wenbin Wang ; Ruiping Wang ; Shiguang Shan ; Xilin Chen
COMMENTS:       Accepted by ECCV 2020
HIGHLIGHT:      Therefore, we argue that a desirable scene graph should be also hierarchically constructed, and introduce a new scheme for modeling scene graph.

49, TITLE:      Two-stream Fusion Model for Dynamic Hand Gesture Recognition using 3D-CNN and 2D-CNN Optical Flow guided Motion Template
http://arxiv.org/abs/2007.08847
AUTHORS:        Debajit Sarma ; V. Kavyasree ; M. K. Bhuyan
COMMENTS:       7 pages, 6 figures, 2 tables. Keywords: Action and gesture recognition, Two-stream fusion model, Optical flow guided motion template (OFMT), 2D and 3D-CNN
HIGHLIGHT:      This work basically proposes a two-stream fusion model for hand gesture recognition and a compact yet efficient motion template based on optical flow.

50, TITLE:      Tackling scalability issues in mining path patterns from knowledge graphs: a preliminary study
http://arxiv.org/abs/2007.08821
AUTHORS:        Pierre Monnin ; Emmanuel Bresso ; Miguel Couceiro ; Malika Smaïl-Tabbone ; Amedeo Napoli ; Adrien Coulet
HIGHLIGHT:      In this paper, we address these issues by proposing a pattern mining approach that relies on a set of constraints (e.g., support or degree thresholds) and the monotonicity property.

51, TITLE:      Neural Architecture Search for Speech Recognition
http://arxiv.org/abs/2007.08818
AUTHORS:        Shoukang Hu ; Xurong Xie ; Shansong Liu ; Mengzhe Geng ; Xunying Liu ; Helen Meng
HIGHLIGHT:      In this paper, a range of neural architecture search (NAS) techniques are used to automatically learn two hyper-parameters that heavily affect the performance and model complexity of state-of-the-art factored time delay neural network (TDNN-F) acoustic models: i) the left and right splicing context offsets; and ii) the dimensionality of the bottleneck linear projection at each hidden layer.

52, TITLE:      Latent Instrumental Variables as Priors in Causal Inference based on Independence of Cause and Mechanism
http://arxiv.org/abs/2007.08812
AUTHORS:        Nataliya Sokolovska ; Pierre-Henri Wuillemin
HIGHLIGHT:      In our contribution, we challenge to reconcile these two research directions.

53, TITLE:      Visual Relation Grounding in Videos
http://arxiv.org/abs/2007.08814
AUTHORS:        Junbin Xiao ; Xindi Shang ; Xun Yang ; Sheng Tang ; Tat-Seng Chua
COMMENTS:       ECCV2020 (spotlight)
HIGHLIGHT:      In this paper, we explore a novel task named visual Relation Grounding in Videos (vRGV).

54, TITLE:      Polarimetric Multi-View Inverse Rendering
http://arxiv.org/abs/2007.08830
AUTHORS:        Jinyu Zhao ; Yusuke Monno ; Masatoshi Okutomi
COMMENTS:       Paper accepted in ECCV 2020
HIGHLIGHT:      In this paper, we propose a novel 3D reconstruction method called Polarimetric Multi-View Inverse Rendering (Polarimetric MVIR) that effectively exploits geometric, photometric, and polarimetric cues extracted from input multi-view color polarization images.

55, TITLE:      Revisiting Rubik's Cube: Self-supervised Learning with Volume-wise Transformation for 3D Medical Image Segmentation
http://arxiv.org/abs/2007.08826
AUTHORS:        Xing Tao ; Yuexiang Li ; Wenhui Zhou ; Kai Ma ; Yefeng Zheng
COMMENTS:       Accepted by MICCAI 2020
HIGHLIGHT:      In this paper, we propose a novel self-supervised learning framework for volumetric medical images.

56, TITLE:      SumGraph: Video Summarization via Recursive Graph Modeling
http://arxiv.org/abs/2007.08809
AUTHORS:        Jungin Park ; Jiyoung Lee ; Ig-Jae Kim ; Kwanghoon Sohn
COMMENTS:       ECCV 2020
HIGHLIGHT:      We propose recursive graph modeling networks for video summarization, termed SumGraph, to represent a relation graph, where frames are regarded as nodes and nodes are connected by semantic relationships among frames.

57, TITLE:      Learning to Combine: Knowledge Aggregation for Multi-Source Domain Adaptation
http://arxiv.org/abs/2007.08801
AUTHORS:        Hang Wang ; Minghao Xu ; Bingbing Ni ; Wenjun Zhang
COMMENTS:       Accepted by ECCV 2020. Code is available at \url{https:github.com/ChrisAllenMing/LtC-MSDA}
HIGHLIGHT:      To mitigate these problems, we propose a Learning to Combine for Multi-Source Domain Adaptation (LtC-MSDA) framework via exploring interactions among domains.

58, TITLE:      Anisotropic Mesh Adaptation for Image Segmentation Based on Mumford-Shah Functional
http://arxiv.org/abs/2007.08696
AUTHORS:        Karrar Abbas ; Xianping Li
COMMENTS:       17 pages, 9 figures
HIGHLIGHT:      In this paper, we consider image segmentation by solving a partial differentiation equation (PDE) model based on the Mumford-Shah functional.

59, TITLE:      Transfer Deep Reinforcement Learning-enabled Energy Management Strategy for Hybrid Tracked Vehicle
http://arxiv.org/abs/2007.08690
AUTHORS:        Xiaowei Guo ; Teng Liu ; Bangbei Tang ; Xiaolin Tang ; Jinwei Zhang ; Wenhao Tan ; Shufeng Jin
COMMENTS:       11 pages, 11 figures
HIGHLIGHT:      This paper proposes an adaptive energy management strategy for hybrid electric vehicles by combining deep reinforcement learning (DRL) and transfer learning (TL).

60, TITLE:      Learn to Propagate Reliably on Noisy Affinity Graphs
http://arxiv.org/abs/2007.08802
AUTHORS:        Lei Yang ; Qingqiu Huang ; Huaiyi Huang ; Linning Xu ; Dahua Lin
COMMENTS:       14 pages, 7 figures, ECCV 2020
HIGHLIGHT:      To overcome these difficulties, we propose a new framework that allows labels to be propagated reliably on large-scale real-world data.

61, TITLE:      Super-Resolution Remote Imaging using Time Encoded Remote Apertures
http://arxiv.org/abs/2007.08667
AUTHORS:        Ji Hyun Nam ; Andreas Velten
HIGHLIGHT:      We show here that it is possible to reconstruct sparse scenes from the temporal profile of the wave-front using only one spatial pixel or a spatial average.

62, TITLE:      Conservative AI and social inequality: Conceptualizing alternatives to bias through social theory
http://arxiv.org/abs/2007.08666
AUTHORS:        Mike Zajko
HIGHLIGHT:      In response to calls for greater interdisciplinary involvement from the social sciences and humanities in the development, governance, and study of artificial intelligence systems, this paper presents one sociologist's view on the problem of algorithmic bias and the reproduction of societal bias.

63, TITLE:      Least squares surface reconstruction on arbitrary domains
http://arxiv.org/abs/2007.08661
AUTHORS:        Dizhong Zhu ; William A P Smith
HIGHLIGHT:      We propose a new method for computing numerical derivatives based on 2D Savitzky-Golay filters and K-nearest neighbour kernels.

64, TITLE:      TUDataset: A collection of benchmark datasets for learning with graphs
http://arxiv.org/abs/2007.08663
AUTHORS:        Christopher Morris ; Nils M. Kriege ; Franka Bause ; Kristian Kersting ; Petra Mutzel ; Marion Neumann
COMMENTS:       ICML 2020 workshop "Graph Representation Learning and Beyond"
HIGHLIGHT:      To address this, we introduce the TUDataset for graph classification and regression.

65, TITLE:      Training with reduced precision of a support vector machine model for text classification
http://arxiv.org/abs/2007.08657
AUTHORS:        Dominik Żurek ; Marcin Pietroń
HIGHLIGHT:      This paper presents the impact of using quantization on the efficiency of multi-class text classification in the training process of a support vector machine (SVM).

66, TITLE:      Technologies for Trustworthy Machine Learning: A Survey in a Socio-Technical Context
http://arxiv.org/abs/2007.08911
AUTHORS:        Ehsan Toreini ; Mhairi Aitken ; Kovila P. L. Coopamootoo ; Karen Elliott ; Vladimiro Gonzalez Zelaya ; Paolo Missier ; Magdalene Ng ; Aad van Moorsel
COMMENTS:       Submitted Version
HIGHLIGHT:      In this paper we provide an overview of technologies that support building trustworthy machine learning systems, i.e., systems whose properties justify that people place trust in them.

67, TITLE:      On the Complexity of Binomialization for Polynomial Differential Equations
http://arxiv.org/abs/2007.08910
AUTHORS:        Mathieu Hemery ; François Fages ; Sylvain Soliman
HIGHLIGHT:      In this paper, we study the theoretical and practical complexities of the bino-mial transformation.

68, TITLE:      Toward Forgetting-Sensitive Referring Expression Generationfor Integrated Robot Architectures
http://arxiv.org/abs/2007.08672
AUTHORS:        Tom Williams ; Torin Johnson ; Will Culpepper ; Kellyn Larson
COMMENTS:       Accepted for (nonarchival) presentation at Advances in Cognitive Systems (ACS) 2020
HIGHLIGHT:      In this work, we computationalize two candidate models of working memory forgetting within a robot cognitive architecture, and demonstrate how they lead to cognitive availability-based differences in generated referring expressions.

69, TITLE:      Deep Small Bowel Segmentation with Cylindrical Topological Constraints
http://arxiv.org/abs/2007.08674
AUTHORS:        Seung Yeon Shin ; Sungwon Lee ; Daniel C. Elton ; James L. Gulley ; Ronald M. Summers
COMMENTS:       Accepted to MICCAI 2020
HIGHLIGHT:      We present a novel method for small bowel segmentation where a cylindrical topological constraint based on persistent homology is applied.

70, TITLE:      Enabling Morally Sensitive Robotic Clarification Requests
http://arxiv.org/abs/2007.08670
AUTHORS:        Ryan Blake Jackson ; Tom Williams
COMMENTS:       Accepted for nonarchival presentation at Advances in Cognitive Systems (ACS) 2020
HIGHLIGHT:      We present a solution to these problems by performing moral reasoning on each potential disambiguation of an ambiguous human utterance and responding accordingly, rather than immediately and naively requesting clarification.

71, TITLE:      Can Learned Frame-Prediction Compete with Block-Motion Compensation for Video Coding?
http://arxiv.org/abs/2007.08922
AUTHORS:        Serkan Sulun ; A. Murat Tekalp
COMMENTS:       Accepted for publication in Springer Journal of Signal, Image and Video Processing
HIGHLIGHT:      Given recent advances in learned video prediction, we investigate whether a simple video codec using a pre-trained deep model for next frame prediction based on previously encoded/decoded frames without sending any motion side information can compete with standard video codecs based on block-motion compensation.

72, TITLE:      Boundary-preserving Mask R-CNN
http://arxiv.org/abs/2007.08921
AUTHORS:        Tianheng Cheng ; Xinggang Wang ; Lichao Huang ; Wenyu Liu
COMMENTS:       17 pages, 8 figures. Accepted by ECCV 2020
HIGHLIGHT:      To remedy these problems, we propose a conceptually simple yet effective Boundary-preserving Mask R-CNN (BMask R-CNN) to leverage object boundary information to improve mask localization accuracy.

73, TITLE:      Consensus-Aware Visual-Semantic Embedding for Image-Text Matching
http://arxiv.org/abs/2007.08883
AUTHORS:        Haoran Wang ; Ying Zhang ; Zhong Ji ; Yanwei Pang ; Lin Ma
COMMENTS:       Accepted by ECCV 2020, Code is publicly available at: https://github.com/BruceW91/CVSE
HIGHLIGHT:      In this paper, we propose a Consensus-aware Visual-Semantic Embedding (CVSE) model to incorporate the consensus information, namely the commonsense knowledge shared between both modalities, into image-text matching.

74, TITLE:      Identification of Tree Species in Japanese Forests based on Aerial Photography and Deep Learning
http://arxiv.org/abs/2007.08907
AUTHORS:        Sarah Kentsch ; Savvas Karatsiolis ; Andreas Kamilaris ; Luca Tomhave ; Maximo Larry Lopez Caceres
COMMENTS:       Proc. of EnviroInfo 2020, Nicosia, Cyprus, September 2020
HIGHLIGHT:      Identification of Tree Species in Japanese Forests based on Aerial Photography and Deep Learning

75, TITLE:      Probabilistic Programming Semantics for Name Generation
http://arxiv.org/abs/2007.08638
AUTHORS:        Marcin Sabok ; Sam Staton ; Dario Stein ; Michael Wolman
COMMENTS:       28 pages, 1 figure
HIGHLIGHT:      We show that quasi-Borel spaces, a model for probabilistic programming, can soundly interpret Stark's $\nu$-calculus, a calculus for name generation.

76, TITLE:      COV-ELM classifier: An Extreme Learning Machine based identification of COVID-19 using Chest-Ray Images
http://arxiv.org/abs/2007.08637
AUTHORS:        Sheetal Rajpal ; Naveen Kumar ; Ankit Rajpal
HIGHLIGHT:      The choice of ELM in this work is based on the fact that ELM significantly shortens the training time with the least interventions required to tune the networks as compared to other neural networks.

77, TITLE:      Superpixel-Guided Label Softening for Medical Image Segmentation
http://arxiv.org/abs/2007.08897
AUTHORS:        Hang Li ; Dong Wei ; Shilei Cao ; Kai Ma ; Liansheng Wang ; Yefeng Zheng
HIGHLIGHT:      In this paper, we propose superpixel-based label softening to tackle the above issue.

78, TITLE:      Task-Level Curriculum Learning for Non-Autoregressive Neural Machine Translation
http://arxiv.org/abs/2007.08772
AUTHORS:        Jinglin Liu ; Yi Ren ; Xu Tan ; Chen Zhang ; Tao Qin ; Zhou Zhao ; Tie-Yan Liu
COMMENTS:       Accepted at IJCAI 2020 Main Track. Sole copyright holder is IJCAI
HIGHLIGHT:      To smooth the shift from AT training to NAT training, in this paper, we introduce semi-autoregressive translation (SAT) as intermediate tasks.

79, TITLE:      Spatial-Spectral Manifold Embedding of Hyperspectral Data
http://arxiv.org/abs/2007.08767
AUTHORS:        Danfeng Hong ; Jing Yao ; Xin Wu ; Jocelyn Chanussot ; Xiao Xiang Zhu
HIGHLIGHT:      Therefore, to reduce the spectral dimensionality effectively and learn more discriminative spectral low-dimensional embedding, in this paper we propose a novel hyperspectral embedding approach by simultaneously considering spatial and spectral information, called spatial-spectral manifold embedding (SSME).

80, TITLE:      A Novel Graph-based Multi-modal Fusion Encoder for Neural Machine Translation
http://arxiv.org/abs/2007.08742
AUTHORS:        Yongjing Yin ; Fandong Meng ; Jinsong Su ; Chulun Zhou ; Zhengyuan Yang ; Jie Zhou ; Jiebo Luo
HIGHLIGHT:      To deal with this issue, in this paper, we propose a novel graph-based multi-modal fusion encoder for NMT.

81, TITLE:      Leveraging both Lesion Features and Procedural Bias in Neuroimaging: An Dual-Task Split dynamics of inverse scale space
http://arxiv.org/abs/2007.08740
AUTHORS:        Xinwei Sun ; Wenjing Han ; Lingjing Hu ; Yuan Yao ; Yizhou Wang
COMMENTS:       Thanks to Xinwei's girlfriend Yue Cao, for her love and support
HIGHLIGHT:      Therefore, in this paper, we propose that the features/voxels in neuroimage data are consist of three orthogonal parts: lesion features, procedural bias, and null features.

82, TITLE:      Channel-wise Autoregressive Entropy Models for Learned Image Compression
http://arxiv.org/abs/2007.08739
AUTHORS:        David Minnen ; Saurabh Singh
COMMENTS:       Published at the IEEE International Conference on Image Processing (ICIP) 2020
HIGHLIGHT:      We introduce two enhancements, channel-conditioning and latent residual prediction, that lead to network architectures with better rate-distortion performance than existing context-adaptive models while minimizing serial processing.

83, TITLE:      Adaptive Task Sampling for Meta-Learning
http://arxiv.org/abs/2007.08735
AUTHORS:        Chenghao Liu ; Zhihao Wang ; Doyen Sahoo ; Yuan Fang ; Kun Zhang ; Steven C. H. Hoi
COMMENTS:       ECCV2020
HIGHLIGHT:      In this paper, we propose an adaptive task sampling method to improve the generalization performance.

84, TITLE:      URIE: Universal Image Enhancement for Visual Recognition in the Wild
http://arxiv.org/abs/2007.08979
AUTHORS:        Taeyoung Son ; Juwon Kang ; Namyup Kim ; Sunghyun Cho ; Suha Kwak
COMMENTS:       17 pages
HIGHLIGHT:      To tackle this issue, we present a Universal and Recognition-friendly Image Enhancement network, dubbed URIE, which is attached in front of existing recognition models and enhances distorted input to improve their performance without retraining them.

85, TITLE:      Proactive Network Maintenance using Fast, Accurate Anomaly Localization and Classification on 1-D Data Series
http://arxiv.org/abs/2007.08752
AUTHORS:        Jingjie Zhu ; Karthik Sundaresan ; Jason Rupe
COMMENTS:       This paper has been accepted by ICPHM 2020
HIGHLIGHT:      We introduce a new algorithm that leverages Deep Convolutional Neural Networks to efficiently and accurately detect anomalies and events on data series, and it reaches 97.82% mean average precision (mAP) in our evaluation.

86, TITLE:      Learning Desirable Matchings From Partial Preferences
http://arxiv.org/abs/2007.09079
AUTHORS:        Hadi Hosseini ; Vijay Menon ; Nisarg Shah ; Sujoy Sikdar
HIGHLIGHT:      Instead of asking the agents to report their complete preferences, our goal is to learn a desirable matching from partial preferences, specifically a matching that is necessarily Pareto optimal (NPO) or necessarily rank-maximal (NRM) under any completion of the partial preferences.

87, TITLE:      Constructing a Family Tree of Ten Indo-European Languages with Delexicalized Cross-linguistic Transfer Patterns
http://arxiv.org/abs/2007.09076
AUTHORS:        Yuanyuan Zhao ; Weiwei Sun ; Xiaojun Wan
HIGHLIGHT:      In this paper, we validate this hypothesis on ten Indo-European languages.

88, TITLE:      Generating Person Images with Appearance-aware Pose Stylizer
http://arxiv.org/abs/2007.09077
AUTHORS:        Siyu Huang ; Haoyi Xiong ; Zhi-Qi Cheng ; Qingzhong Wang ; Xingran Zhou ; Bihan Wen ; Jun Huan ; Dejing Dou
COMMENTS:       Appearing at IJCAI 2020. The code is available at https://github.com/siyuhuang/PoseStylizer
HIGHLIGHT:      In this paper, we present a novel end-to-end framework to generate realistic person images based on given person poses and appearances.

89, TITLE:      Sequential Explanations with Mental Model-Based Policies
http://arxiv.org/abs/2007.09028
AUTHORS:        Arnold YS Yeung ; Shalmali Joshi ; Joseph Jay Williams ; Frank Rudzicz
COMMENTS:       Accepted into ICML 2020 Workshop on Human Interpretability in Machine Learning (Spotlight)
HIGHLIGHT:      We apply a reinforcement learning framework which emulates this format by providing explanations based on the explainee's current mental model.

90, TITLE:      Parameterized Complexity of Scheduling Chains of Jobs with Delays
http://arxiv.org/abs/2007.09023
AUTHORS:        Hans L. Bodlaender ; Marieke van der Wegen
HIGHLIGHT:      In this paper, we consider the parameterized complexity of the following scheduling problem.

91, TITLE:      Knowledge-Based Video Question Answering with Unsupervised Scene Descriptions
http://arxiv.org/abs/2007.08751
AUTHORS:        Noa Garcia ; Yuta Nakashima
HIGHLIGHT:      Inspired by this behaviour, we design ROLL, a model for knowledge-based video story question answering that leverages three crucial aspects of movie understanding: dialog comprehension, scene reasoning, and storyline recalling.

92, TITLE:      GMNet: Graph Matching Network for Large Scale Part Semantic Segmentation in the Wild
http://arxiv.org/abs/2007.09073
AUTHORS:        Umberto Michieli ; Edoardo Borsato ; Luca Rossi ; Pietro Zanuttigh
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this work, we propose a novel framework combining higher object-level context conditioning and part-level spatial relationships to address the task.

93, TITLE:      Hybrid Discriminative-Generative Training via Contrastive Learning
http://arxiv.org/abs/2007.09070
AUTHORS:        Hao Liu ; Pieter Abbeel
COMMENTS:       Code: https://github.com/lhao499/HDGE
HIGHLIGHT:      In this paper we show that through the perspective of hybrid discriminative-generative training of energy-based models we can make a direct connection between contrastive learning and supervised learning.

94, TITLE:      Standing on the Shoulders of Giants: Hardware and Neural Architecture Co-Search with Hot Start
http://arxiv.org/abs/2007.09087
AUTHORS:        Weiwen Jiang ; Lei Yang ; Sakyasingha Dasgupta ; Jingtong Hu ; Yiyu Shi
COMMENTS:       13 pages
HIGHLIGHT:      In this paper, we propose a novel framework, namely HotNAS, that starts from a "hot" state based on a set of existing pre-trained models (a.k.a. model zoo) to avoid lengthy training time.

95, TITLE:      Towards an Automated SOAP Note: Classifying Utterancesfrom Medical Conversations
http://arxiv.org/abs/2007.08749
AUTHORS:        Benjamin Schloss ; Sandeep Konam
COMMENTS:       22 pages,1 figure
HIGHLIGHT:      In the current paper, we focus on two tasks: classifying utterances from medical conversations according to (i)the SOAP section and (ii) the speaker role, both fundamental building blocks along the path towards an end-to-end, automated SOAP note for medical conversations.

96, TITLE:      Online Invariance Selection for Local Feature Descriptors
http://arxiv.org/abs/2007.08988
AUTHORS:        Rémi Pautrat ; Viktor Larsson ; Martin R. Oswald ; Marc Pollefeys
COMMENTS:       27 pages, Accepted at ECCV 2020 (Oral)
HIGHLIGHT:      To be invariant, or not to be invariant: that is the question formulated in this work about local descriptors.

97, TITLE:      Backdoor Learning: A Survey
http://arxiv.org/abs/2007.08745
AUTHORS:        Yiming Li ; Baoyuan Wu ; Yong Jiang ; Zhifeng Li ; Shu-Tao Xia
COMMENTS:       11 pages. A curated list of backdoor learning resources in this paper is presented in the Github Repo (https://github.com/THUYimingLi/backdoor-learning-resources). We will try our best to continuously maintain the repo
HIGHLIGHT:      This paper presents the first comprehensive survey on the backdoor learning.

98, TITLE:      TopoAL: An Adversarial Learning Approach for Topology-Aware Road Segmentation
http://arxiv.org/abs/2007.09084
AUTHORS:        Subeesh Vasu ; Mateusz Kozinski ; Leonardo Citraro ; Pascal Fua
HIGHLIGHT:      To address this issue, we introduce an Adversarial Learning (AL) strategy tailored for our purposes.

99, TITLE:      Multi-Stage Influence Function
http://arxiv.org/abs/2007.09081
AUTHORS:        Hongge Chen ; Si Si ; Yang Li ; Ciprian Chelba ; Sanjiv Kumar ; Duane Boning ; Cho-Jui Hsieh
HIGHLIGHT:      In this paper, we develop a multi-stage influence function score to track predictions from a finetuned model all the way back to the pretraining data.

100, TITLE:     A Technical Report for VIPriors Image Classification Challenge
http://arxiv.org/abs/2007.08722
AUTHORS:        Zhipeng Luo ; Ge Li ; Zhiguang Zhang
COMMENTS:       ECCV2020,VIPriors Image Classification Challenge
HIGHLIGHT:      In this challenge, the difficulty is how to train the model from scratch without any pretrained weight.

101, TITLE:     Hyperparameter Selection for Offline Reinforcement Learning
http://arxiv.org/abs/2007.09055
AUTHORS:        Tom Le Paine ; Cosmin Paduraru ; Andrea Michi ; Caglar Gulcehre ; Konrad Zolna ; Alexander Novikov ; Ziyu Wang ; Nando de Freitas
HIGHLIGHT:      Therefore, in this work, we focus on \textit{offline hyperparameter selection}, i.e. methods for choosing the best policy from a set of many policies trained using different hyperparameters, given only logged data.

102, TITLE:     Multi-scale Interactive Network for Salient Object Detection
http://arxiv.org/abs/2007.09062
AUTHORS:        Youwei Pang ; Xiaoqi Zhao ; Lihe Zhang ; Huchuan Lu
COMMENTS:       Accepted by CVPR 2020
HIGHLIGHT:      In this paper, we propose the aggregate interaction modules to integrate the features from adjacent levels, in which less noise is introduced because of only using small up-/down-sampling rates.

103, TITLE:     Region-based Non-local Operation for Video Classification
http://arxiv.org/abs/2007.09033
AUTHORS:        Guoxi Huang ; Adrian G. Bors
HIGHLIGHT:      This paper presents region-based non-local operation (RNL), a family of self-attention mechanisms, which can directly capture long-range dependencies without a deep stack of local operations.

104, TITLE:     Learning to Discretely Compose Reasoning Module Networks for Video Captioning
http://arxiv.org/abs/2007.09049
AUTHORS:        Ganchao Tan ; Daqing Liu ; Meng Wang ; Zheng-Jun Zha
COMMENTS:       Accepted at IJCAI 2020 Main Track. Sole copyright holder is IJCAI. Code is available at https://github.com/tgc1997/RMN
HIGHLIGHT:      In this paper, we propose a novel visual reasoning approach for video captioning, named Reasoning Module Networks (RMN), to equip the existing encoder-decoder framework with the above reasoning capacity.

105, TITLE:     Advances in Deep Learning for Hyperspectral Image Analysis--Addressing Challenges Arising in Practical Imaging Scenarios
http://arxiv.org/abs/2007.08592
AUTHORS:        Xiong Zhou ; Saurabh Prasad
COMMENTS:       Published as a chapter in Hyperspectral Image Analysis. Advances in Computer Vision and Pattern Recognition
HIGHLIGHT:      In this chapter, we will review recent advances in the community that leverage deep learning for robust hyperspectral image analysis despite these unique challenges -- specifically, we will review unsupervised, semi-supervised and active learning approaches to image analysis, as well as transfer learning approaches for multi-source (e.g. multi-sensor, or multi-temporal) image analysis.

106, TITLE:     SqueezeFacePoseNet: Lightweight Face Verification Across Different Poses for Mobile Platforms
http://arxiv.org/abs/2007.08566
AUTHORS:        Fernando Alonso-Fernandez ; Javier Barrachina ; Kevin Hernandez-Diaz ; Josef Bigun
HIGHLIGHT:      In this paper, we adapt the lightweight SqueezeNet model, of just 4.4MB, to effectively provide cross-pose face recognition.

107, TITLE:     Integer factorization and Riemann's hypothesis: Why two-item joint replenishment is hard
http://arxiv.org/abs/2007.09045
AUTHORS:        Andreas S. Schulz ; Claudio Telha
COMMENTS:       20 pages, 4 figures
HIGHLIGHT:      Joint replenishment problems are a fundamental model in inventory management, manufacturing, and logistics that capture these effects.



==========Updates to Previous Papers==========
1, TITLE:       Reducing the Sim-to-Real Gap for Event Cameras
http://arxiv.org/abs/2003.09078
AUTHORS:        Timo Stoffregen ; Cedric Scheerlinck ; Davide Scaramuzza ; Tom Drummond ; Nick Barnes ; Lindsay Kleeman ; Robert Mahony
COMMENTS:       Updated to final submitted version (ECCV 2020). Updated results to reflect updated models + updated evaluation metric for optic flow (more fair, so our model shows less performance improvements compared to previously reported) + updated values for image reconstruction after finding bug in evaluating SoTA (failing to normalize input tensors for inference).g
HIGHLIGHT:      To address this, we present a new High Quality Frames (HQF) dataset, containing events and ground truth frames from a DAVIS240C that are well-exposed and minimally motion-blurred.

2, TITLE:       Gradient Descent over Metagrammars for Syntax-Guided Synthesis
http://arxiv.org/abs/2007.06677
AUTHORS:        Nicolas Chan ; Elizabeth Polgreen ; Sanjit A. Seshia
COMMENTS:       5 pages, SYNT 2020
HIGHLIGHT:      In this work, we speculate this default grammar could be improved upon substantially. We build sets of rules, or metagrammars, for constructing grammars, and perform a gradient descent over these metagrammars aiming to find a metagrammar which solves more benchmarks and on average faster.

3, TITLE:       RobustScanner: Dynamically Enhancing Positional Clues for Robust Text Recognition
http://arxiv.org/abs/2007.07542
AUTHORS:        Xiaoyu Yue ; Zhanghui Kuang ; Chenhao Lin ; Hongbin Sun ; Wayne Zhang
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      To suppress such side-effect, we propose a novel position enhancement branch, and dynamically fuse its outputs with those of the decoder attention module for scene text recognition.

4, TITLE:       Accelerating Reinforcement Learning Agent with EEG-based Implicit Human Feedback
http://arxiv.org/abs/2006.16498
AUTHORS:        Duo Xu ; Mohit Agarwal ; Ekansh Gupta ; Faramarz Fekri ; Raghupathy Sivakumar
HIGHLIGHT:      In this work, we investigate capturing human's intrinsic reactions as implicit (and natural) feedback through EEG in the form of error-related potentials (ErrP), providing a natural and direct way for humans to improve the RL agent learning.

5, TITLE:       Learning Surrogates via Deep Embedding
http://arxiv.org/abs/2007.00799
AUTHORS:        Yash Patel ; Tomas Hodan ; Jiri Matas
COMMENTS:       ECCV 2020 camera-ready version
HIGHLIGHT:      This paper proposes a technique for training a neural network by minimizing a surrogate loss that approximates the target evaluation metric, which may be non-differentiable.

6, TITLE:       Improving 3D Object Detection through Progressive Population Based Augmentation
http://arxiv.org/abs/2004.00831
AUTHORS:        Shuyang Cheng ; Zhaoqi Leng ; Ekin Dogus Cubuk ; Barret Zoph ; Chunyan Bai ; Jiquan Ngiam ; Yang Song ; Benjamin Caine ; Vijay Vasudevan ; Congcong Li ; Quoc V. Le ; Jonathon Shlens ; Dragomir Anguelov
COMMENTS:       Accepted at ECCV 2020
HIGHLIGHT:      In this work, we present the first attempt to automate the design of data augmentation policies for 3D object detection.

7, TITLE:       An Ensemble of Epoch-wise Empirical Bayes for Few-shot Learning
http://arxiv.org/abs/1904.08479
AUTHORS:        Yaoyao Liu ; Bernt Schiele ; Qianru Sun
HIGHLIGHT:      In this paper, we propose to meta-learn the ensemble of epoch-wise empirical Bayes models (E3BM) to achieve robust predictions.

8, TITLE:       Efficient Semantic Video Segmentation with Per-frame Inference
http://arxiv.org/abs/2002.11433
AUTHORS:        Yifan Liu ; Chunhua Shen ; Changqian Yu ; Jingdong Wang
COMMENTS:       Accepted to Proc. Eur. Conf. Computer Vision (ECCV), 2020
HIGHLIGHT:      In this work, we process efficient semantic video segmentation in a per-frame fashion during the inference process.

9, TITLE:       GANHopper: Multi-Hop GAN for Unsupervised Image-to-Image Translation
http://arxiv.org/abs/2002.10102
AUTHORS:        Wallace Lira ; Johannes Merz ; Daniel Ritchie ; Daniel Cohen-Or ; Hao Zhang
COMMENTS:       To be presented at ECCV 2020. Code is available at https://github.com/wallacemplira/ganhopper
HIGHLIGHT:      We introduce GANHopper, an unsupervised image-to-image translation network that transforms images gradually between two domains, through multiple hops.

10, TITLE:      StRDAN: Synthetic-to-Real Domain Adaptation Network for Vehicle Re-Identification
http://arxiv.org/abs/2004.12032
AUTHORS:        Sangrok Lee ; Eunsoo Park ; Hongsuk Yi ; Sang Hun Lee
COMMENTS:       7 pages, 2 figures, CVPR Workshop Paper (Revised)
HIGHLIGHT:      Therefore, we propose a synthetic-to-real domain adaptation network (StRDAN) framework, which can be trained with inexpensive large-scale synthetic and real data to improve performance.

11, TITLE:      Wavelet-Based Dual-Branch Network for Image Demoireing
http://arxiv.org/abs/2007.07173
AUTHORS:        Lin Liu ; Jianzhuang Liu ; Shanxin Yuan ; Gregory Slabaugh ; Ales Leonardis ; Wengang Zhou ; Qi Tian
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      In this paper, we design a wavelet-based dual-branch network (WDNet) with a spatial attention mechanism for image demoireing.

12, TITLE:      70 years of machine learning in geoscience in review
http://arxiv.org/abs/2006.13311
AUTHORS:        Jesper Sören Dramsch
COMMENTS:       36 pages, 17 figures, book chapter
HIGHLIGHT:      Regarding geoscience, the review has a bias towards geophysics but aims to strike a balance with geochemistry, geostatistics, and geology, however excludes remote sensing, as this would exceed the scope.

13, TITLE:      Cooking Is All About People: Comment Classification On Cookery Channels Using BERT and Classification Models (Malayalam-English Mix-Code)
http://arxiv.org/abs/2007.04249
AUTHORS:        Subramaniam Kazhuparambil ; Abhishek Kaushik
COMMENTS:       Rectified typos
HIGHLIGHT:      In this work, we have evaluated top-performing classification models for classifying comments which are a mix of different combinations of English and Malayalam (only English, only Malayalam and Mix of English and Malayalam).

14, TITLE:      Curriculum DeepSDF
http://arxiv.org/abs/2003.08593
AUTHORS:        Yueqi Duan ; Haidong Zhu ; He Wang ; Li Yi ; Ram Nevatia ; Leonidas J. Guibas
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper, we design a "shape curriculum" for learning continuous Signed Distance Function (SDF) on shapes, namely Curriculum DeepSDF.

15, TITLE:      Iterative Label Improvement: Robust Training by Confidence Based Filtering and Dataset Partitioning
http://arxiv.org/abs/2002.02705
AUTHORS:        Christian Haase-Schütz ; Rainer Stal ; Heinz Hertlein ; Bernhard Sick
HIGHLIGHT:      To alleviate this issue, we propose a novel meta training and labelling scheme that is able to use inexpensive unlabelled data by taking advantage of the generalization power of deep neural networks.

16, TITLE:      Vision Meets Drones: Past, Present and Future
http://arxiv.org/abs/2001.06303
AUTHORS:        Pengfei Zhu ; Longyin Wen ; Dawei Du ; Xiao Bian ; Qinghua Hu ; Haibin Ling
COMMENTS:       arXiv admin note: text overlap with arXiv:1804.07437
HIGHLIGHT:      After that, we describe our VisDrone dataset, which is captured over various urban/suburban areas of 14 different cities across China from North to South.

17, TITLE:      Learning to Take Directions One Step at a Time
http://arxiv.org/abs/1812.01874
AUTHORS:        Qiyang Hu ; Adrian Wälchli ; Tiziano Portenier ; Matthias Zwicker ; Paolo Favaro
HIGHLIGHT:      We present a method to generate a video sequence given a single image.

18, TITLE:      How to trust unlabeled data? Instance Credibility Inference for Few-Shot Learning
http://arxiv.org/abs/2007.08461
AUTHORS:        Yikai Wang ; Li Zhang ; Yuan Yao ; Yanwei Fu
COMMENTS:       Journal extension of arXiv:2003.11853 that appears in CVPR 2020
HIGHLIGHT:      Different to prior arts that leverage meta-learning or data augmentation strategies to alleviate this extremely data-scarce problem, this paper presents a statistical approach, dubbed Instance Credibility Inference to exploit the support of unlabeled instances for few-shot visual recognition.

19, TITLE:      U-Net Based Architecture for an Improved Multiresolution Segmentation in Medical Images
http://arxiv.org/abs/2007.08238
AUTHORS:        Simindokht Jahangard ; Mohammad Hossein Zangooei ; Maysam Shahedi
HIGHLIGHT:      In this study, our objective is to improve the multi-resolution image segmentation performance of U-Net architecture.

20, TITLE:      VisualEchoes: Spatial Image Representation Learning through Echolocation
http://arxiv.org/abs/2005.01616
AUTHORS:        Ruohan Gao ; Changan Chen ; Ziad Al-Halah ; Carl Schissler ; Kristen Grauman
COMMENTS:       Appears in ECCV 2020
HIGHLIGHT:      We explore the spatial cues contained in echoes and how they can benefit vision tasks that require spatial reasoning.

21, TITLE:      Geometric Wavelet Scattering Networks on Compact Riemannian Manifolds
http://arxiv.org/abs/1905.10448
AUTHORS:        Michael Perlmutter ; Feng Gao ; Guy Wolf ; Matthew Hirn
COMMENTS:       35 pages; 3 figures; 2 tables; v3: Revisions based on reviewer comments
HIGHLIGHT:      Inspired by recent interest in geometric deep learning, which aims to generalize convolutional neural networks to manifold and graph-structured domains, we define a geometric scattering transform on manifolds.

22, TITLE:      Synthesis in Uclid5
http://arxiv.org/abs/2007.06760
AUTHORS:        Federico Mora ; Kevin Cheang ; Elizabeth Polgreen ; Sanjit A. Seshia
HIGHLIGHT:      We describe an integration of program synthesis into Uclid5, a formal modelling and verification tool. We use the integration to generate 25 program synthesis benchmarks with simple, known solutions that are out of reach of current synthesis engines, and we release the benchmarks to the community.

23, TITLE:      Federated Visual Classification with Real-World Data Distribution
http://arxiv.org/abs/2003.08082
AUTHORS:        Tzu-Ming Harry Hsu ; Hang Qi ; Matthew Brown
HIGHLIGHT:      In this work, we characterize the effect these real-world data distributions have on distributed learning, using as a benchmark the standard Federated Averaging (FedAvg) algorithm. To do so, we introduce two new large-scale datasets for species and landmark classification, with realistic per-user data splits that simulate real-world edge learning scenarios.

24, TITLE:      Faster than FAST: GPU-Accelerated Frontend for High-Speed VIO
http://arxiv.org/abs/2003.13493
AUTHORS:        Balazs Nagy ; Philipp Foehn ; Davide Scaramuzza
COMMENTS:       IEEE International Conference on Intelligent Robots and Systems (IROS), 2020. Open-source implementation available at https://github.com/uzh-rpg/vilib
HIGHLIGHT:      Our second contribution introduces an enhanced FAST feature detector that applies the aforementioned non-maxima suppression method.

25, TITLE:      SpatialSim: Recognizing Spatial Configurations of Objects with Graph Neural Networks
http://arxiv.org/abs/2004.04546
AUTHORS:        Laetitia Teodorescu ; Katja Hofmann ; Pierre-Yves Oudeyer
HIGHLIGHT:      In this paper we make two key contributions.

26, TITLE:      Dynamic Node Embeddings from Edge Streams
http://arxiv.org/abs/1904.06449
AUTHORS:        John Boaz Lee ; Giang Nguyen ; Ryan A. Rossi ; Nesreen K. Ahmed ; Eunyee Koh ; Sungchul Kim
COMMENTS:       IEEE Transactions on Emerging Topics in Computational Intelligence (TETIC)
HIGHLIGHT:      In this work, we propose using the notion of temporal walks for learning dynamic embeddings from temporal networks.

27, TITLE:      DHP: Differentiable Meta Pruning via HyperNetworks
http://arxiv.org/abs/2003.13683
AUTHORS:        Yawei Li ; Shuhang Gu ; Kai Zhang ; Luc Van Gool ; Radu Timofte
COMMENTS:       ECCV camera-ready. Code is available at https://github.com/ofsoundof/dhp
HIGHLIGHT:      To circumvent this problem, this paper introduces a differentiable pruning method via hypernetworks for automatic network pruning.

28, TITLE:      Defocus Deblurring Using Dual-Pixel Data
http://arxiv.org/abs/2005.00305
AUTHORS:        Abdullah Abuolaim ; Michael S. Brown
COMMENTS:       Camera-ready version
HIGHLIGHT:      We propose an effective defocus deblurring method that exploits data available on dual-pixel (DP) sensors found on most modern cameras.

29, TITLE:      Context-Gated Convolution
http://arxiv.org/abs/1910.05577
AUTHORS:        Xudong Lin ; Lin Ma ; Wei Liu ; Shih-Fu Chang
COMMENTS:       ECCV 2020 camera ready version with appendix
HIGHLIGHT:      Motivated by this, we propose one novel Context-Gated Convolution (CGC) to explicitly modify the weights of convolutional layers adaptively under the guidance of global context.

30, TITLE:      EAGLE: Large-scale Vehicle Detection Dataset inReal-World Scenarios using Aerial Imagery
http://arxiv.org/abs/2007.06124
AUTHORS:        Seyed Majid Azimi ; Reza Bahmanyar ; Corenin Henry ; Franz Kurz
HIGHLIGHT:      To address this issue, we introduce EAGLE (oriEnted vehicle detection using Aerial imaGery in real-worLd scEnarios), a large-scale dataset for multi-class vehicle detection with object orientation information in aerial imagery.

31, TITLE:      Graph-Based Social Relation Reasoning
http://arxiv.org/abs/2007.07453
AUTHORS:        Wanhua Li ; Yueqi Duan ; Jiwen Lu ; Jianjiang Feng ; Jie Zhou
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper, we propose a simpler, faster, and more accurate method named graph relational reasoning network (GR2N) for social relation recognition.

32, TITLE:      Text Recognition -- Real World Data and Where to Find Them
http://arxiv.org/abs/2007.03098
AUTHORS:        Klára Janoušková ; Jiri Matas ; Lluis Gomez ; Dimosthenis Karatzas
COMMENTS:       10 pages
HIGHLIGHT:      We present a method for exploiting weakly annotated images to improve text extraction pipelines.

33, TITLE:      LIIR at SemEval-2020 Task 12: A Cross-Lingual Augmentation Approach for Multilingual Offensive Language Identification
http://arxiv.org/abs/2005.03695
AUTHORS:        Erfan Ghadery ; Marie-Francine Moens
HIGHLIGHT:      For other languages we propose a cross-lingual augmentation approach in order to enrich training data and we use Multilingual BERT to obtain sentence representations.

34, TITLE:      Efficient Attention Mechanism for Visual Dialog that can Handle All the Interactions between Multiple Inputs
http://arxiv.org/abs/1911.11390
AUTHORS:        Van-Quang Nguyen ; Masanori Suganuma ; Takayuki Okatani
COMMENTS:       Accepted to ECCV 2020, 14 pages. Slight change in title
HIGHLIGHT:      In this paper, we present a neural architecture named Light-weight Transformer for Many Inputs (LTMI) that can efficiently deal with all the interactions between multiple such inputs in visual dialog.

35, TITLE:      Approximation spaces of deep neural networks
http://arxiv.org/abs/1905.01208
AUTHORS:        Rémi Gribonval ; Gitta Kutyniok ; Morten Nielsen ; Felix Voigtlaender
HIGHLIGHT:      We study the expressivity of deep neural networks.

36, TITLE:      Synchronization of Deterministic Visibly Push-Down Automata
http://arxiv.org/abs/2005.01374
AUTHORS:        Henning Fernau ; Petra Wolf
HIGHLIGHT:      We generalize the concept of synchronizing words for finite automata, which map all states of the automata to the same state, to deterministic visibly push-down automata.

37, TITLE:      Dual Past and Future for Neural Machine Translation
http://arxiv.org/abs/2007.07728
AUTHORS:        Jianhao Yan ; Fandong Meng ; Jie Zhou
HIGHLIGHT:      In this paper, we present a novel dual framework that leverages both source-to-target and target-to-source NMT models to provide a more direct and accurate supervision signal for the Past and Future modules.

38, TITLE:      SF-Net: Single-Frame Supervision for Temporal Action Localization
http://arxiv.org/abs/2003.06845
AUTHORS:        Fan Ma ; Linchao Zhu ; Yi Yang ; Shengxin Zha ; Gourab Kundu ; Matt Feiszli ; Zheng Shou
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper, we study an intermediate form of supervision, i.e., single-frame supervision, for temporal action localization (TAL).

39, TITLE:      Dual Mixup Regularized Learning for Adversarial Domain Adaptation
http://arxiv.org/abs/2007.03141
AUTHORS:        Yuan Wu ; Diana Inkpen ; Ahmed El-Roby
COMMENTS:       This paper has been accepted by ECCV 2020
HIGHLIGHT:      In order to alleviate the above issues, we propose a dual mixup regularized learning (DMRL) method for UDA, which not only guides the classifier in enhancing consistent predictions in-between samples, but also enriches the intrinsic structures of the latent space.

40, TITLE:      Component Order Connectivity in Directed Graphs
http://arxiv.org/abs/2007.06896
AUTHORS:        J. Bang-Jensen ; E. Eiben ; G. Gutin ; M. Wahlstrom ; A. Yeo
HIGHLIGHT:      $ We study parametered complexity of DCOC for general and semicomplete digraphs with the following parameters: $k, \ell,\ell+k$ and $n-\ell$.

41, TITLE:      Real-time high speed motion prediction using fast aperture-robust event-driven visual flow
http://arxiv.org/abs/1811.11135
AUTHORS:        Himanshu Akolkar ; SioHoi Ieng ; Ryad Benosman
COMMENTS:       Pre-print version, 12 pages, 12 figures. Accepted in IEEE tPAMI July 2020
HIGHLIGHT:      In this paper, we propose a novel multi-scale plane fitting based visual flow algorithm that is robust to the aperture problem and also computationally fast and efficient.

42, TITLE:      BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models
http://arxiv.org/abs/2003.11142
AUTHORS:        Jiahui Yu ; Pengchong Jin ; Hanxiao Liu ; Gabriel Bender ; Pieter-Jan Kindermans ; Mingxing Tan ; Thomas Huang ; Xiaodan Song ; Ruoming Pang ; Quoc Le
COMMENTS:       Accepted in ECCV 2020
HIGHLIGHT:      In this work, we propose BigNAS, an approach that challenges the conventional wisdom that post-processing of the weights is necessary to get good prediction accuracies.

43, TITLE:      Making Robots Draw A Vivid Portrait In Two Minutes
http://arxiv.org/abs/2005.05526
AUTHORS:        Fei Gao ; Jingjie Zhu ; Zeyuan Yu ; Peng Li ; Tao Wang
COMMENTS:       7 pages, 7 figures; accepted by IROS2020
HIGHLIGHT:      Besides, we propose a componential sparsity constraint to reduce the number of brush-strokes over insignificant areas.

44, TITLE:      Image Classification in the Dark using Quanta Image Sensors
http://arxiv.org/abs/2006.02026
AUTHORS:        Abhiram Gnanasambandam ; Stanley H. Chan
COMMENTS:       Published in the 16th European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      In this paper, we present a new low-light image classification solution using Quanta Image Sensors (QIS).

45, TITLE:      Guiding Deep Molecular Optimization with Genetic Exploration
http://arxiv.org/abs/2007.04897
AUTHORS:        Sungsoo Ahn ; Junsu Kim ; Hankook Lee ; Jinwoo Shin
HIGHLIGHT:      In this paper, we propose genetic expert-guided learning (GEGL), a simple yet novel framework for training a deep neural network (DNN) to generate highly-rewarding molecules.

46, TITLE:      Probabilistic Future Prediction for Video Scene Understanding
http://arxiv.org/abs/2003.06409
AUTHORS:        Anthony Hu ; Fergal Cotter ; Nikhil Mohan ; Corina Gurau ; Alex Kendall
COMMENTS:       Accepted as a conference paper at ECCV 2020
HIGHLIGHT:      To model the stochasticity of the future, we introduce a conditional variational approach which minimises the divergence between the present distribution (what could happen given what we have seen) and the future distribution (what we observe actually happens).

47, TITLE:      Deep Decomposition Learning for Inverse Imaging Problems
http://arxiv.org/abs/1911.11028
AUTHORS:        Dongdong Chen ; Mike E. Davies
COMMENTS:       To appear in ECCV 2020
HIGHLIGHT:      In this paper, inspired by the geometry that data can be decomposed by two components from the null-space of the forward operator and the range space of its pseudo-inverse, we train neural networks to learn the two components and therefore learn the decomposition, i.e. we explicitly reformulate the neural network layers as learning range-nullspace decomposition functions with reference to the layer inputs, instead of learning unreferenced functions.

48, TITLE:      Machine Intelligence at the Edge with Learning Centric Power Allocation
http://arxiv.org/abs/1911.04922
AUTHORS:        Shuai Wang ; Yik-Chung Wu ; Minghua Xia ; Rui Wang ; H. Vincent Poor
COMMENTS:       14 figures, 15 pages, to appear in IEEE Transactions on Wireless Communications
HIGHLIGHT:      To this end, this paper proposes learning centric power allocation (LCPA), which provides a new perspective on radio resource allocation in learning driven scenarios.

49, TITLE:      Synchronization under Dynamic Constraints
http://arxiv.org/abs/1910.01935
AUTHORS:        Petra Wolf
HIGHLIGHT:      We present three attempts to model constraints of these kinds on the order in which the states of an automaton are transitioned by a synchronizing word.

50, TITLE:      Train Your Data Processor: Distribution-Aware and Error-Compensation Coordinate Decoding for Human Pose Estimation
http://arxiv.org/abs/2007.05887
AUTHORS:        Feiyu Yang ; Zhan Song ; Zhenzhong Xiao ; Yu Chen ; Zhe Pan ; Min Zhang ; Min Xue ; Yaoyang Mo ; Yao Zhang ; Guoxiong Guan ; Beibei Qian
COMMENTS:       Improve the state-of-the-art of COCO keypoint detection challenge by 1-2 AP. Project page: https://github.com/fyang235/DAEC
HIGHLIGHT:      Serving as a model-agnostic plug-in, DAEC learns its decoding strategy from training data and remarkably improves the performance of a variety of state-of-the-art human pose estimation models with negligible extra computation.

51, TITLE:      Dispersed Exponential Family Mixture VAEs for Interpretable Text Generation
http://arxiv.org/abs/1906.06719
AUTHORS:        Wenxian Shi ; Hao Zhou ; Ning Miao ; Lei Li
COMMENTS:       Accepted by ICML 2020
HIGHLIGHT:      In this paper, we find that mode-collapse is a general problem for VAEs with exponential family mixture priors.

52, TITLE:      Counting Query Answers over a DL-Lite Knowledge Base (extended version)
http://arxiv.org/abs/2005.05886
AUTHORS:        Diego Calvanese ; Julien Corman ; Davide Lanti ; Simon Razniewski
COMMENTS:       Extended version of an article published at IJCAI 2020
HIGHLIGHT:      In this paper we focus on counting answers over a Knowledge Base (KB), which may be viewed as a database enriched with background knowledge about the domain under consideration.

53, TITLE:      Explaining Image Classifiers using Statistical Fault Localization
http://arxiv.org/abs/1908.02374
AUTHORS:        Youcheng Sun ; Hana Chockler ; Xiaowei Huang ; Daniel Kroening
HIGHLIGHT:      In this paper, we show that statistical fault localization (SFL) techniques from software engineering deliver high quality explanations of the outputs of DNNs, where we define an explanation as a minimal subset of features sufficient for making the same decision as for the original input.

54, TITLE:      Iris Presentation Attack Detection: Where Are We Now?
http://arxiv.org/abs/2006.13252
AUTHORS:        Aidan Boyd ; Zhaoyuan Fang ; Adam Czajka ; Kevin W. Bowyer
COMMENTS:       Under revision for Pattern Recognition Letters
HIGHLIGHT:      This work presents an overview of the most important advances in the area of iris presentation attack detection published in recent two years.

55, TITLE:      Empirical Analysis of Zipf's Law, Power Law, and Lognormal Distributions in Medical Discharge Reports
http://arxiv.org/abs/2003.13352
AUTHORS:        Juan C Quiroz ; Liliana Laranjo ; Catalin Tufanaru ; Ahmet Baki Kocaballi ; Dana Rezazadegan ; Shlomo Berkovsky ; Enrico Coiera
COMMENTS:       Reduced word count to 3000, moved methods details to appendices
HIGHLIGHT:      We examined 20,000 medical discharge reports from the MIMIC-III dataset.

56, TITLE:      Structured3D: A Large Photo-realistic Dataset for Structured 3D Modeling
http://arxiv.org/abs/1908.00222
AUTHORS:        Jia Zheng ; Junfei Zhang ; Jing Li ; Rui Tang ; Shenghua Gao ; Zihan Zhou
COMMENTS:       Accepted to ECCV 2020. Project website: https://structured3d-dataset.org
HIGHLIGHT:      In this paper, we present a new synthetic dataset, Structured3D, with the aim of providing large-scale photo-realistic images with rich 3D structure annotations for a wide spectrum of structured 3D modeling tasks.

57, TITLE:      On the I/O complexity of hybrid algorithms for Integer Multiplication
http://arxiv.org/abs/1912.08045
AUTHORS:        Lorenzo De Stefani
COMMENTS:       arXiv admin note: substantial text overlap with arXiv:1904.12804
HIGHLIGHT:      We present an $\Omega\left(\left(n/\max\{M,n_0\}\right)^{\log_k \left(2k-1\right)}\left(\max\{1,n_0/M\}\right)^2M\right)$ lower bound for the I/O complexity of a class of "uniform, non-stationary" hybrid algorithms, where $n_0$ denotes the threshold size of sub-problems which are computed using standard algorithms with algebraic complexity $\Omega\left(n^2\right)$.

58, TITLE:      Design and Interpretation of Universal Adversarial Patches in Face Detection
http://arxiv.org/abs/1912.05021
AUTHORS:        Xiao Yang ; Fangyun Wei ; Hongyang Zhang ; Jun Zhu
HIGHLIGHT:      We propose new optimization-based approaches to automatic design of universal adversarial patches for varying goals of the attack, including scenarios in which true positives are suppressed without introducing false positives.

59, TITLE:      RANSAC-Flow: generic two-stage image alignment
http://arxiv.org/abs/2004.01526
AUTHORS:        Xi Shen ; François Darmon ; Alexei A. Efros ; Mathieu Aubry
COMMENTS:       Accepted to ECCV 2020 as a spotlight. Project page: http://imagine.enpc.fr/~shenx/RANSAC-Flow/
HIGHLIGHT:      Our main insight is that parametric and non-parametric alignment methods have complementary strengths.

60, TITLE:      Cortical-inspired Wilson-Cowan-type equations for orientation-dependent contrast perception modelling
http://arxiv.org/abs/1910.06808
AUTHORS:        Marcelo Bertalmío ; Luca Calatroni ; Valentina Franceschi ; Benedetta Franceschiello ; Dario Prandi
COMMENTS:       This is the revised extended invited journal version of the SSVM 2019 conference proceeding arXiv:1812.07425
HIGHLIGHT:      We consider the evolution model proposed in [9, 6] to describe illusory contrast perception phenomena induced by surrounding orientations.

61, TITLE:      Attention Guided Anomaly Localization in Images
http://arxiv.org/abs/1911.08616
AUTHORS:        Shashanka Venkataramanan ; Kuan-Chuan Peng ; Rajat Vikram Singh ; Abhijit Mahalanobis
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      Without the need of anomalous training images, we propose Convolutional Adversarial Variational autoencoder with Guided Attention (CAVGA), which localizes the anomaly with a convolutional latent variable to preserve the spatial information.

62, TITLE:      Predicting Citywide Crowd Flows in Irregular Regions Using Multi-View Graph Convolutional Networks
http://arxiv.org/abs/1903.07789
AUTHORS:        Junkai Sun ; Junbo Zhang ; Qiaofei Li ; Xiuwen Yi ; Yuxuan Liang ; Yu Zheng
COMMENTS:       12 pages, 13 figures, 5 tables. Published in IEEE TKDE, Date of Publication: Jul. 13, 2020
HIGHLIGHT:      In this paper, we formulate crowd flow forecasting in irregular regions as a spatio-temporal graph (STG) prediction problem in which each node represents a region with time-varying flows.

63, TITLE:      NE-LP: Normalized Entropy and Loss Prediction based Sampling for Active Learning in Chinese Word Segmentation on EHRs
http://arxiv.org/abs/1908.08419
AUTHORS:        Tingting Cai ; Zhiyuan Ma ; Hong Zheng ; Yangming Zhou
COMMENTS:       submitted to Neural Computing and Applications
HIGHLIGHT:      In this paper, we follow the trend and present an active learning method for CWS in EHRs.

64, TITLE:      Unsupervised Domain Adaptation for Semantic Segmentation of NIR Images through Generative Latent Search
http://arxiv.org/abs/2006.08696
AUTHORS:        Prashant Pandey ; Aayush Kumar Tyagi ; Sameer Ambekar ; Prathosh AP
COMMENTS:       ECCV 2020 [Spotlight]
HIGHLIGHT:      We propose a method for target-independent segmentation where the 'nearest-clone' of a target image in the source domain is searched and used as a proxy in the segmentation network trained only on the source domain.

65, TITLE:      Neural Belief Reasoner
http://arxiv.org/abs/1909.04719
AUTHORS:        Haifeng Qian
HIGHLIGHT:      This paper proposes a new generative model called neural belief reasoner (NBR).

66, TITLE:      Discriminative Feature Alignment: Improving Transferability of Unsupervised Domain Adaptation by Gaussian-guided Latent Alignment
http://arxiv.org/abs/2006.12770
AUTHORS:        Jing Wang ; Jiahong Chen ; Jianzhe Lin ; Leonid Sigal ; Clarence W. de Silva
COMMENTS:       15 pages, 12 figures
HIGHLIGHT:      In this study, we focus on the unsupervised domain adaptation problem where an approximate inference model is to be learned from a labeled data domain and expected to generalize well to an unlabeled data domain.

67, TITLE:      Dissociable neural representations of adversarially perturbed images in convolutional neural networks and the human brain
http://arxiv.org/abs/1812.09431
AUTHORS:        Chi Zhang ; Xiaohan Duan ; Linyuan Wang ; Yongli Li ; Bin Yan ; Guoen Hu ; Ruyuan Zhang ; Li Tong
HIGHLIGHT:      Here, we leverage adversarial noise (AN) and adversarial interference (AI) images to quantify the consistency between neural representations and perceptual outcomes in the two systems.

68, TITLE:      Invisible Backdoor Attacks on Deep Neural Networks via Steganography and Regularization
http://arxiv.org/abs/1909.02742
AUTHORS:        Shaofeng Li ; Minhui Xue ; Benjamin Zi Hao Zhao ; Haojin Zhu ; Xinpeng Zhang
HIGHLIGHT:      In this paper, we create covert and scattered triggers for backdoor attacks, invisible backdoors, where triggers can fool both DNN models and human inspection.

69, TITLE:      Universal Composability is Secure Compilation
http://arxiv.org/abs/1910.08634
AUTHORS:        Marco Patrignani ; Riad S. Wahby ; Robert Künnemann
HIGHLIGHT:      This paper outlines the connection between universal composability and robust compilation, the latest of secure compilation theories.

70, TITLE:      Neural Wireframe Renderer: Learning Wireframe to Image Translations
http://arxiv.org/abs/1912.03840
AUTHORS:        Yuan Xue ; Zihan Zhou ; Xiaolei Huang
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper, we bridge the information gap by generating photo-realistic rendering of indoor scenes from wireframe models in an image translation framework.

71, TITLE:      Counterfactual Vision-and-Language Navigation via Adversarial Path Sampling
http://arxiv.org/abs/1911.07308
AUTHORS:        Tsu-Jui Fu ; Xin Eric Wang ; Matthew Peterson ; Scott Grafton ; Miguel Eckstein ; William Yang Wang
COMMENTS:       ECCV 2020 (spotlight)
HIGHLIGHT:      We propose an adversarial-driven counterfactual reasoning model that can consider effective conditions instead of low-quality augmented data.

72, TITLE:      The Devil is in the Details: Self-Supervised Attention for Vehicle Re-Identification
http://arxiv.org/abs/2004.06271
AUTHORS:        Pirazh Khorramshahi ; Neehar Peri ; Jun-cheng Chen ; Rama Chellappa
COMMENTS:       This work has been accepted European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      In this paper, we present Self-supervised Attention for Vehicle Re-identification (SAVER), a novel approach to effectively learn vehicle-specific discriminative features.

73, TITLE:      Towards a quantum-inspired proof for IP = PSPACE
http://arxiv.org/abs/1912.11611
AUTHORS:        Ayal Green ; Guy Kindler ; Yupan Liu
COMMENTS:       10 pages
HIGHLIGHT:      We explore quantum-inspired interactive proof systems where the prover is limited.

74, TITLE:      Neural Design Network: Graphic Layout Generation with Constraints
http://arxiv.org/abs/1912.09421
AUTHORS:        Hsin-Ying Lee ; Lu Jiang ; Irfan Essa ; Phuong B Le ; Haifeng Gong ; Ming-Hsuan Yang ; Weilong Yang
COMMENTS:       European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      We propose a method for design layout generation that can satisfy user-specified constraints.

75, TITLE:      Explanation Augmented Feedback in Human-in-the-Loop Reinforcement Learning
http://arxiv.org/abs/2006.14804
AUTHORS:        Lin Guan ; Mudit Verma ; Subbarao Kambhampati
HIGHLIGHT:      To address this, we present EXPAND (Explanation Augmented Feedback) which allows for explanatory information to be given as saliency maps from the human in addition to the binary feedback.

76, TITLE:      Comparing to Learn: Surpassing ImageNet Pretraining on Radiographs By Comparing Image Representations
http://arxiv.org/abs/2007.07423
AUTHORS:        Hong-Yu Zhou ; Shuang Yu ; Cheng Bian ; Yifan Hu ; Kai Ma ; Yefeng Zheng
COMMENTS:       MICCAI 2020 early accept; Code and pretrained models available at this http URL http://github.com/funnyzhou/C2L_MICCAI2020
HIGHLIGHT:      To bridge this gap, we propose a new pretraining method which learns from 700k radiographs given no manual annotations.

77, TITLE:      Learning Manifolds from Non-stationary Streaming Data
http://arxiv.org/abs/1804.08833
AUTHORS:        Suchismit Mahapatra ; Varun Chandola
COMMENTS:       27 pages, 9 figures
HIGHLIGHT:      We present theoretical results to show that the quality of a manifold asymptotically converges as the size of data increases.

78, TITLE:      A Closest Point Proposal for MCMC-based Probabilistic Surface Registration
http://arxiv.org/abs/1907.01414
AUTHORS:        Dennis Madsen ; Andreas Morel-Forster ; Patrick Kahr ; Dana Rahbani ; Thomas Vetter ; Marcel Lüthi
HIGHLIGHT:      We propose to view non-rigid surface registration as a probabilistic inference problem.

79, TITLE:      Controllable Image Synthesis via SegVAE
http://arxiv.org/abs/2007.08397
AUTHORS:        Yen-Chi Cheng ; Hsin-Ying Lee ; Min Sun ; Ming-Hsuan Yang
COMMENTS:       ECCV 2020. Project page: https://yccyenchicheng.github.io/SegVAE/ Code: https://github.com/yccyenchicheng/SegVAE
HIGHLIGHT:      In this work, we specifically target at generating semantic maps given a label-set consisting of desired categories.

80, TITLE:      Dependency Stochastic Boolean Satisfiability: A Logical Formalism for NEXPTIME Decision Problems with Uncertainty
http://arxiv.org/abs/1911.04112
AUTHORS:        Nian-Ze Lee ; Jie-Hong R. Jiang
COMMENTS:       10 pages, 3 figures
HIGHLIGHT:      With the theoretical foundations paved in this work, our results may encourage DSSAT solver development to enable potential broad applications.

81, TITLE:      Placepedia: Comprehensive Place Understanding with Multi-Faceted Annotations
http://arxiv.org/abs/2007.03777
AUTHORS:        Huaiyi Huang ; Yuqi Zhang ; Qingqiu Huang ; Zhengkui Guo ; Ziwei Liu ; Dahua Lin
COMMENTS:       To appear in ECCV 2020. Dataset is available at: https://hahehi.github.io/placepedia.html
HIGHLIGHT:      In this work, we contribute Placepedia, a large-scale place dataset with more than 35M photos from 240K unique places.

82, TITLE:      Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
http://arxiv.org/abs/2006.09882
AUTHORS:        Mathilde Caron ; Ishan Misra ; Julien Mairal ; Priya Goyal ; Piotr Bojanowski ; Armand Joulin
HIGHLIGHT:      In this paper, we propose an online algorithm, SwAV, that takes advantage of contrastive methods without requiring to compute pairwise comparisons.

83, TITLE:      Efficient Document Exchange and Error Correcting Codes with Asymmetric Information
http://arxiv.org/abs/2007.00870
AUTHORS:        Kuan Cheng ; Xin Li
HIGHLIGHT:      In this paper we study whether asymmetric partial information can help in these two problems.

84, TITLE:      SideInfNet: A Deep Neural Network for Semi-Automatic Semantic Segmentation with Side Information
http://arxiv.org/abs/2002.02634
AUTHORS:        Jing Yu Koh ; Duc Thanh Nguyen ; Quang-Trung Truong ; Sai-Kit Yeung ; Alexander Binder
COMMENTS:       ECCV 2020
HIGHLIGHT:      Inspired by the practicality and applicability of the semi-automatic approach, this paper proposes a novel deep neural network architecture, namely SideInfNet that effectively integrates features learnt from images with side information extracted from user annotations.

85, TITLE:      Information Extraction of Clinical Trial Eligibility Criteria
http://arxiv.org/abs/2006.07296
AUTHORS:        Yitong Tseo ; M. I. Salkola ; Ahmed Mohamed ; Anuj Kumar ; Freddy Abnousi
COMMENTS:       4 pages
HIGHLIGHT:      In this paper, we investigate an information extraction (IE) approach for grounding criteria from trials in ClinicalTrials(dot)gov to a shared knowledge base.

86, TITLE:      GANSpace: Discovering Interpretable GAN Controls
http://arxiv.org/abs/2004.02546
AUTHORS:        Erik Härkönen ; Aaron Hertzmann ; Jaakko Lehtinen ; Sylvain Paris
HIGHLIGHT:      This paper describes a simple technique to analyze Generative Adversarial Networks (GANs) and create interpretable controls for image synthesis, such as change of viewpoint, aging, lighting, and time of day.

87, TITLE:      Structure-Aware Human-Action Generation
http://arxiv.org/abs/2007.01971
AUTHORS:        Ping Yu ; Yang Zhao ; Chunyuan Li ; Junsong Yuan ; Changyou Chen
COMMENTS:       accepted by ECCV 2020
HIGHLIGHT:      To overcome this issue, we propose a variant of GCNs to leverage the powerful self-attention mechanism to adaptively sparsify a complete action graph in the temporal space.

88, TITLE:      Event-based Asynchronous Sparse Convolutional Networks
http://arxiv.org/abs/2003.09148
AUTHORS:        Nico Messikommer ; Daniel Gehrig ; Antonio Loquercio ; Davide Scaramuzza
HIGHLIGHT:      In this work, we present a general framework for converting models trained on synchronous image-like event representations into asynchronous models with identical output, thus directly leveraging the intrinsic asynchronous and sparse nature of the event data.

89, TITLE:      Evolving Normalization-Activation Layers
http://arxiv.org/abs/2004.02967
AUTHORS:        Hanxiao Liu ; Andrew Brock ; Karen Simonyan ; Quoc V. Le
HIGHLIGHT:      Here we propose to design them using an automated approach.

90, TITLE:      G-LBM:Generative Low-dimensional Background Model Estimation from Video Sequences
http://arxiv.org/abs/2003.07335
AUTHORS:        Behnaz Rezaei ; Amirreza Farnoosh ; Sarah Ostadabbas
HIGHLIGHT:      In this paper, we propose a computationally tractable and theoretically supported non-linear low-dimensional generative model to represent real-world data in the presence of noise and sparse outliers.

91, TITLE:      The Effect of Moderation on Online Mental Health Conversations
http://arxiv.org/abs/2005.09225
AUTHORS:        David Wadden ; Tal August ; Qisheng Li ; Tim Althoff
COMMENTS:       Accepted as a full paper at ICWSM 2021. 13 pages, 12 figures, 3 tables
HIGHLIGHT:      In this work, we leveraged a natural experiment, occurring across 200,000 messages from 7,000 conversations hosted on a mental health mobile application, to evaluate the effects of moderation on online mental health discussions.

92, TITLE:      Generating Videos of Zero-Shot Compositions of Actions and Objects
http://arxiv.org/abs/1912.02401
AUTHORS:        Megha Nawhal ; Mengyao Zhai ; Andreas Lehrmann ; Leonid Sigal ; Greg Mori
COMMENTS:       Accepted at ECCV'20; Project Page: https://www.sfu.ca/~mnawhal/projects/zs_hoi_generation.html
HIGHLIGHT:      In this paper we develop methods for generating such videos -- making progress toward addressing the important, open problem of video generation in complex scenes.

93, TITLE:      Type-based Declassification for Free
http://arxiv.org/abs/1905.00922
AUTHORS:        Minh Ngo ; David A. Naumann ; Tamara Rezk
COMMENTS:       The short version of this paper is accepted in ICFEM 2020. 100 pages
HIGHLIGHT:      This work provides a study to demonstrate the potential of using off-the-shelf programming languages and their theories to build sound language-based-security tools.