Skip to content

Commit

Permalink
Update index.html
Browse files Browse the repository at this point in the history
  • Loading branch information
@Gynjn
Gynjn authored Aug 7, 2024
1 parent 7d692bd commit 83dff45
Showing 1 changed file with 0 additions and 341 deletions.
341 changes: 0 additions & 341 deletions index.html
View file
Original file line number Diff line number Diff line change
@@ -1,342 +1 @@
<!DOCTYPE html>
<html>
!
<head>
<meta charset="utf-8">
<title>Codec NeRF</title>
<meta name="viewport" content="width=device-width, initial-scale=1">

<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=G-PYVRSFMDRL"></script>
<script>
window.dataLayer = window.dataLayer || [];

function gtag() {
dataLayer.push(arguments);
}

gtag('js', new Date());

gtag('config', 'G-PYVRSFMDRL');
</script>

<link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet">

<link rel="stylesheet" href="./static/css/bulma.min.css">
<link rel="stylesheet" href="./static/css/bulma-carousel.min.css">
<link rel="stylesheet" href="./static/css/bulma-slider.min.css">
<link rel="stylesheet" href="./static/css/fontawesome.all.min.css">
<link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
<link rel="stylesheet" href="./static/css/index.css">
<!-- <link rel="icon" href="./static/images/favicon.svg"> -->

<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
<script defer src="./static/js/fontawesome.all.min.js"></script>
<script src="./static/js/bulma-carousel.min.js"></script>
<script src="./static/js/bulma-slider.min.js"></script>
<script src="./static/js/index.js"></script>
<script src="./static/js/script.js"></script>

</head>

<body>
<section class="hero">sadsadsadsad
<div class="hero-body">asdsadasd
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<h1 class="title is-1 publication-title">CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis</h1>
<div class="is-size-5 publication-authors">
<span class="author-block">
<A href="https://github.com/Gynjn">Gyeongjin Kang</A><sup>*</sup>,</span>
<span class="author-block">
<A href="https://github.com/Younggeun-L">Younggeun Lee</A><sup>*</sup>,</span>
<span class="author-block">
<A href="https://github.com/ohsngjun">Seungjun Oh</A>, and </span>
<span class="author-block">
<A href="https://silverbottlep.github.io/">Eunbyung Park</A></span>

</div>

<div class="is-size-5 publication-authors">
<span class="author-block">Sungkyunkwan University</span> <br>
<span class="author-block"><sup>*</sup>Equal contribution</span>
</div>
<div class="column has-text-centered">
<span class="link-block">
<a href="https://arxiv.org/abs/2404.04913"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
</span>
<span>Paper</span>
</a>
</span>
<!-- Code Link. -->
<span class="link-block">
<a href="https://github.com/Gynjn/Codec-NeRF"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Code</span>
</a>
</span>
</div>

</div>
</div>
</div>
</div>
</div>
</section>


<section class="hero is-light is-small">
<div class="hero-body">
<div class="container">
<div class=" has-text-centered">
<h2 class="title is-3">Performance Comparison on Objaverse Dataset</h2>
<div id="results-carousel" class="carousel results-carousel">
<div class="item item-steve">
<video poster="" id="steve" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/t1.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-shiba">
<video poster="" id="shiba" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/t2.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-fullbody">
<video poster="" id="fullbody" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/t3.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-blueshirt">
<video poster="" id="blueshirt" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/t4.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-blueshirt">
<video poster="" id="blueshirt" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/t5.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-blueshirt">
<video poster="" id="blueshirt" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/t6.mp4"
type="video/mp4">
</video>
</div>
</div>
</div>
</div>
</section>

<section class="hero is-light is-small">
<div class="hero-body">
<div class="container">
<div class=" has-text-centered">
<h2 class="title is-3">Performance Comparison on ShapeNet Dataset</h2>
<div id="results-carousel" class="carousel results-carousel">
<div class="item item-steve">
<video poster="" id="steve" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/c1.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-chair-tp">
<video poster="" id="chair-tp" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/cc1.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-shiba">
<video poster="" id="shiba" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/c2.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-fullbody">
<video poster="" id="fullbody" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/cc2.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-toby">
<video poster="" id="toby" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/c3.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-toby">
<video poster="" id="toby" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/cc3.mp4"
type="video/mp4">
</video>
</div>
</div>
</div>
</div>
</div>
</section>

<section class="section">
<div class="container is-max-desktop">
<!-- Abstract. -->
<!-- abstract -->
<p class="title is-3 mt-5 has-text-centered"> Abstract </p>
<p class="content is-size-6 has-text-left">
Neural Radiance Fields (NeRF) have achieved huge success in effectively capturing and representing 3D objects and scenes.
However, several factors have impeded its further proliferation as next-generation 3D media.
To establish a ubiquitous presence in everyday media formats, such as images and videos, it is imperative to devise a solution that effectively fulfills three key objectives: fast encoding and decoding time, compact model sizes, and high-quality renderings.
Despite significant advancements, a comprehensive algorithm that adequately addresses all objectives has yet to be fully realized.
In this work, we present <b>CodecNeRF</b>, a neural codec for NeRF representations, consisting of a novel encoder and decoder architecture that can generate a NeRF representation in a single forward pass.
Furthermore, inspired by the recent parameter-efficient finetuning approaches, we develop a novel finetuning method to efficiently adapt the generated NeRF representations to a new test instance, leading to high-quality image renderings and compact code sizes. The proposed CodecNeRF, a newly suggested encoding-decoding-finetuning pipeline for NeRF, achieved unprecedented compression performance of more than 150x and 20x reduction in encoding time while maintaining (or improving) the image quality on widely used 3D object datasets, such as ShapeNet and Objaverse.
</p>
<!--/ Abstract. -->

<!-- Paper video. -->
</div>
</section>




<section class="section">
<div class="container is-max-desktop">
<!-- Concurrent Work. -->
<div class="columns is-centered has-text-centered">
<div class="column is-full-width">
<h2 class="title is-3">Codec NeRF Pipeline</h2>
<div class="content has-text-justified">
<img src="./static/images/overall_arch.png">
<p align="justify">Given N input images from different viewpoints, the goal is to produce a NeRF representation (multi-resolution triplanes).
First, a 2D image feature extractor module processes all input images and generates 2D feature maps for each input image.
hen, the unproject and aggregation module lifts the 2D features to 3D features and aggregates the unprojected 3D features into a single 3D feature.
The 3D feature is further processed by axis-aligned average pooling along each axis, resulting in three 2D features.
These 2D features are used to generate multi-resolution triplanes and finally, we perform the volumetric rendering to render an image using MLP.
Furthermore, 2D features are compressed by the compression module, producing the minimal sizes of the codes to be transmitted. The entire pipeline is differentiable, and we train end-to-end to optimize all learnable parameters.</p>
<img src="./static/images/PEFT.png">
<p align="justify">Similar to other NeRF generalization models, our approach can also
leverage fine-tuning of NeRF representations to enhance visual quality for new
scenes during testing time. We propose to adapt parameter-efficient finetuning (PEFT) in our test time optimization.
We first generate multi-resolution triplanes using
multi-view test images, and train only the triplanes and decoder in a efficient way.
Also, we leverage neural compression methods that have demonstrated efficacy in
image and video domains to seek to achieve the optimal compression rate. We adopt a fully factorized density model to our proposed
parameter-efficient finetuning of the triplanes and MLP.
</p>
</div>
</div>
</div>
</div>
</section>

<section class="section">
<div class="container is-max-desktop">
<!-- Concurrent Work. -->
<div class="columns is-centered has-text-centered">
<div class="column is-full-width">
<h2 class="title is-3">Entropy Coding visualization</h2>
<img src="./static/images/featmap.png" style="width:100%">
<h2 class="title is-4" style="margin-top: 50px;">w/ Entropy Coding <strong style="color: red;">(top)</strong> &nbsp; vs. &nbsp; w/o Entropy Coding <strong style="color: red;">(bottom)</strong></h2>
<img src="./static/images/thist.jpg" style="width:100%">

<p align="justify">We visualize the delta feature maps across finetuning iterations using our entropy coding method.
The feature maps following the entropy coding, eliminate unnecessary components across the different resolutions, and sequentially get a high
compression ratio resulting from quantization. The histogram shows the progress of compression during finetuning stage.
</p>

</div>
</div>
</div>
</section>

<section class="section">
<div class="container is-max-desktop">
<!-- Concurrent Work. -->
<div class="columns is-centered has-text-centered">
<div class="column is-full-width">
<h2 class="title is-3">Encoding time and compression performance</h2>
<img src="./static/images/fixed.jpg" style="width:100%">

<p align="justify">For the quantitative metrics, we report the standard
image quality measurements, PSNR, SSIM, and MS-SSIM. We also measure the
storage requirements of the representation to show the compression performance
of our method. Our parameter efficient methods show that only with 150x memory we can store or transmit the 3D representation with better quality compared
to the baseline method. Due to the improved generalization capability of our
method, we outperform the per-scene optimization-based baseline, Triplanes,
in novel view synthesis. Remarkably, the PEFT method showed comparable or
better performance across the data with fewer trainable parameters.
</p>

</div>
</div>
</div>
</section>

<script>
new BeforeAfter({
id: '#example1'
});
new BeforeAfter({
id: '#example2'
});
new BeforeAfter({
id: '#example4'
});
new BeforeAfter({
id: '#example5'
});

</script>

<section class="section" id="BibTeX">
<div class="container is-max-desktop content">
<h2 class="title">BibTeX</h2>
<pre><code>
@misc{kang2024codecnerf,
title={CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis},
author={Gyeongjin Kang and Younggeun Lee and Eunbyung Park},
year={2024},
eprint={2404.04913},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
</code></pre>
</div>
</section>

<footer class="footer">
<div class="container">
<div class="content has-text-centered">
<!-- <a class="icon-link" href="https://arxiv.org/abs/2401.00834"> -->
<a class="icon-link" href="https://arxiv.org/abs/2404.04913">
<i class="fas fa-file-pdf"></i>
</a>
<a class="icon-link" href="https://github.com/Gynjn/Codec-NeRF" class="external-link" disabled>
<i class="fab fa-github"></i>
</a>
</div>
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>
This website borrowed template from <a href="https://github.com/nerfies/nerfies.github.io">nerfies</a>.
</p>
</div>
</div>
</div>
</div>
</footer>

</body>

</html>

1 comment on commit 83dff45

@Gynjn
Copy link
Owner Author

@Gynjn Gynjn commented on 83dff45 Aug 7, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

d

Please sign in to comment.