Skip to content

Commit

Permalink
Merge pull request #18 from shubhMaheshwari/main
Browse files Browse the repository at this point in the history 
Updating website to BIGE
  • Loading branch information
@shubhMaheshwari
shubhMaheshwari authored Dec 6, 2024
2 parents ec886f1 + 4d10c12 commit 903ea98
Show file tree
Hide file tree
Showing 6 changed files with 1,817 additions and 31 deletions.
90 changes: 60 additions & 30 deletions docs/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,12 +3,12 @@
<head>
<meta charset="utf-8">
<meta name="description"
content="Style-based controllable motion generation">
content="BIGE : Biomechanics-informed GenAI for Exercise Science">
<meta name="keywords" content="Generative AI, Stable diffusio, Animation">
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta http-equiv="Permissions-Policy" content="interest-cohort=()">

<title>Style-based controllable motion generation</title>
<title>BIGE : Biomechanics-informed GenAI for Exercise Science</title>

<!-- Global site tag (gtag.js) - Google Analytics -->
<!-- <script async src="https://www.googletagmanager.com/gtag/js?id=G-PYVRSFMDRL"></script>
Expand Down Expand Up @@ -50,23 +50,39 @@
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<h1 class="title is-1 publication-title">Style-based controllable motion generation</h1>
<h1 class="title is-1 publication-title">BIGE : Biomechanics-informed GenAI for Exercise Science</h1>
<div class="is-size-5 publication-authors">
<span class="author-block"><a href="https://www.linkedin.com/in/paninib">Panini Bhamidipati</a>,</span>
<span class="author-block"><a href="https://shubhmaheshwari.github.io/">Shubh Maheshwari</a>,</span>
<span class="author-block"><a href="https://www.linkedin.com/in/anwesh-mohanty">Anwesh Mohanty</a>,</span>
<span class="author-block"><a href="https://www.linkedin.com/in/anwesh-mohanty">Yadi Cao</a>,</span>
<span class="author-block"><a href="https://roseyu.com/">Rose Yu</a></span>
</div>

<div class="is-size-5 publication-authors">
<span class="author-block">Department of Computer Science</span><br>
<span class="author-block">UC San Diego</span>
<span class="author-block">
&emsp; &emsp; &emsp; Rose Spatio-Temporal Lab<sup>1</sup>, &emsp; &emsp;
</span>
<span class="author-block">
Cardiac Mechanics Research Group <sup>2</sup>
</span>
<br>
<span class="author-block">
Department of Computer Science &emsp; &emsp;
</span>
<span class="author-block">
Department of Bioengineering
</span>
<br>
</div>
<div class="is-size-5 publication-authors">
<span class="author-block">University of California San Diego</span>
</div>

<div class="column has-text-centered">
<div class="publication-links">
<!-- PDF Link. -->
<span class="link-block">
<a href=""
<a href="./static/pdfs/L4DC_2025_paper_177.pdf"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
Expand All @@ -75,7 +91,7 @@ <h1 class="title is-1 publication-title">Style-based controllable motion generat
</a>
</span>
<span class="link-block">
<a href=""
<a href="./static/pdfs/Supplementary_L4DC_2025_paper_177.pdf"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
Expand All @@ -85,12 +101,12 @@ <h1 class="title is-1 publication-title">Style-based controllable motion generat
</span>
<!-- Video Link. -->
<span class="link-block">
<a href=""
<a href="https://docs.google.com/presentation/d/15Cpc11clPP0p72jBd7Zoe1e7_FpdCYRn3uXChIdGpzY/edit?usp=sharing"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fa fa-solid fa-image"></i>
</span>
<span>Poster</span>
<span>Slidies</span>
</a>
</span>
<!-- Code Link. -->
Expand All @@ -105,7 +121,7 @@ <h1 class="title is-1 publication-title">Style-based controllable motion generat
</span>
<!-- Dataset Link. -->
<span class="link-block">
<a href="#" onclick="alert('Coming soon')"
<a href="./ucsd-opencap-dataset.html"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="far fa-images"></i>
Expand All @@ -121,15 +137,22 @@ <h1 class="title is-1 publication-title">Style-based controllable motion generat
</div>
</section>


<section class="hero teaser">
<div class="container is-max-desktop">
<div class="hero-body">
<video id="teaser" autoplay muted loop playsinline height="100%">
<!-- <video id="teaser" autoplay muted loop playsinline height="100%">
<source src="./static/videos/demo.mp4"
type="video/mp4">
</video>
</video> -->
<img src="./static/images/Sports_Analytic_Mockup_1_V4.png"
class="interpolation-image"
alt="Interpolation end reference image."/>

<h2 class="subtitle has-text-centered">
A novel constraint composition for motion diffusion to recommend clinically optimal body poses.
<b> BIGE</b> is a framework for generative models to adhere to clinician-defined constraints.
<br>
To generate realistic motion, our method uses a biomechanically informed surrogate model to guide the generation process..
</h2>
</div>
</div>
Expand Down Expand Up @@ -159,32 +182,44 @@ <h2 class="title is-3">Summary</h2>
<div class="column is-four-fifths">
<div class="content has-text-justified">



<p>
Recommending clinically optimal body motion can significantly improve an athlete's performance. The goal of our work is to suggest optimal body poses an athlete should perform. To distinguish between optimal and sub-optimal pose we use the Movement Competency Screen (MCS) to guide the generative model to produce motion with high fidelity and diversity.
Proper movements enhance mobility, coordination, and muscle activation, which are crucial for
performance, injury prevention, and overall fitness. However, traditional simulation tools rely
on strong modeling assumptions, are difficult to set up and computationally expensive. On the
other hand, generative AI approaches provide efficient alternatives to motion generation. But they
often lack physiological relevance and do not incorporate biomechanical constraints, limiting their
practical applications in sports and exercise science. To address these limitations.
</p>

<ul>
<li>
<p>
We prosose a novel constraint composition formulation to incorporate multiple constraints like MCS Score and text prompt.
We propose a novel framework, BIGE, that combines bio-mechanically meaningful scoring metrics with generative modeling.
</p>
</li>
<li>
<p>
Furthermore, we release a new dataset (OpenCap Dataset) encompassing more than 3000 high-quality human motion samples from 13 different primitive movements like Squats, Pull-ups, and Push-ups with their corresponding MCS Score labeled by experts.
BIGE integrates a differentiable surrogate model for muscle activation to reverse optimize the latent space of the generative model
</p>
</li>
<li>
<p>
Our experiments show qualitative and quantitative improvement over baselines by incorporating constrain composition.
Enables the retrieval of physiologically valid motions through targeted search.
</p>
</li>
<li>
<p>
Through extensive experiments on squat exercise data, our framework demonstrates superior performance in generating diverse, physically plausible motions while maintaining high fidelity to clinician-defined objectives compared to existing approaches.
</p>
</li>
</ul>
</div>
</div>
</div>
<!--/ Abstract. -->

<!-- Paper video. -->
<!-- <div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
Expand Down Expand Up @@ -351,7 +386,7 @@ <h2 class="title is-3 has-text-centered">Temporal Segmentation</h2>
</section>


<section class="section">
<!-- <section class="section">
<div class="container is-max-desktop has-text-centered">
<h2 class="title is-3"> Generating Style Tags </h2>
<div class="content has-text-justified">
Expand All @@ -378,30 +413,26 @@ <h2 class="title is-3"> Generating Style Tags </h2>
</div>
</div>
</section>

</section> -->

<!--
<section class="section hero">
<div class="container is-max-desktop">
<div class="columns is-centered">
<!-- Animation. -->
<div class="columns is-centered">
<div class="column is-full-width">
<h2 class="title is-3 has-text-centered">Comparisions</h2>
<!-- Static. -->
<h3 class="title is-4 has-text-centered">Constraint composition</h3>
<div class="content has-text-justified">
<img src="./static/images/kit-comparision.png"
class="interpolation-image"
alt="Interpolation end reference image."/>
</div>
<br/>
<!--/ Static. -->
<!-- Motion Skeleton. -->
<h3 class="title is-4 has-text-centered"> Stylized generatation</h3>
<div class="content has-text-justified">
<img src="./static/images/opecap-comparision.png"
Expand All @@ -411,19 +442,18 @@ <h3 class="title is-4 has-text-centered"> Stylized generatation</h3>
Compared to motion skeleton extracted from other methods, by incorporating structural cues, ours is more effective at embedding skeleton from incomplete mesh sequence.
</p>
</div>
<!--/ Motion Skeleton. -->
</div>
<!--/ Matting. -->
</div>
</section>
</section> -->

<section class="section is-light" id="BibTeX">
<div class="container is-max-desktop content">
<h2 class="title">BibTeX</h2>
<pre><code>@article{,
author = {Bhamidipati, Panini and Maheshwari, Shubh and Yu, Rose},
title = {Style-based controllable motion generation},
title = {BIGE : Biomechanics-informed GenAI for Exercise Science},
journal = {},
year = {2024},
}</code></pre>
Expand Down
Binary file added docs/static/images/Sports_Analytic_Mockup_1_V4.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading

0 comments on commit 903ea98

Please sign in to comment.