Найден корректный robots.txt
Файл управления сканированием сайта роботами
?
Robots.txt указывает поисковым роботам, какие страницы сканировать, а какие — нет. Ошибки в файле могут случайно закрыть важные разделы от индексации.
Robots.txt настроен корректно. Размер файла: 23333 байт. Загружен за: 1сек.
Проверяемая страница не запрещена в robots.txt.
Цепочка редиректов для файла robots.txt:
http://lsd-3d.com/robots.txt
302 Found
https://princeton-computational-imaging.github.io/LSD-3D/
200 OK
Показать содержимое robots.txt
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="description"
content="Large-Scale 3D Driving Scene Generation with Geometry Grounding.">
<meta name="keywords" content="Scene Generation, Gaussian Splatting, GenAI, Gen AI, 3D Generation, Autonomous Driving">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding</title>
<link rel="icon" type="image/png" href="https://light.princeton.edu/wp-content/uploads/2023/05/pci-only-seal-dark.png">
<link rel="shortcut icon" type="image/png" href="https://light.princeton.edu/wp-content/uploads/2023/05/pci-only-seal-dark.png">
<!-- 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>
<!-- Preload critical resources -->
<link rel="preload" href="static/css/styles.css" as="style">
<link rel="preload" href="static/js/script_optimized.js" as="script">
<link rel="preconnect" href="https://cdnjs.cloudflare.com">
<link rel="preconnect" href="https://sparkjs.dev">
<!-- <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="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.0.0/css/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/styles.css">
<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
<script src="static/js/bulma-carousel.min.js"></script>
</head>
<body>
<section class="hero">
<!-- <video class="hero-background-video" autoplay muted loop playsinline>
<source src="assets/video/zoom_4x4_high_res_compressed.mp4" type="video/mp4">
</video> -->
<div class="hero-body">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<h1 class="title is-1 publication-title">LSD-3D: Large-Scale 3D Driving Scene <br> Generation with Geometry Grounding</h1>
<p class="is-size-4" style="margin-bottom: 20px; font-weight: 600; color: #555;">AAAI-26</p>
<div class="is-size-5 publication-authors">
<!-- <span class="author-block"><a href="https://ostjul.com">Julian Ost</a><sup>*,1</sup>,</span>
<span class="author-block"><a href="https://andrearama.github.io">Andrea Ramazzina</a><sup>,*2</sup>,</span>
<span class="author-block"><a href="https://amogh7joshi.github.io">Amogh Joshi</a><sup>*,1</sup>,</span>
<span class="author-block"><a href="">Maximillian Bömer</a><sup></sup>,</span>
<span class="author-block"><a href="http://mariobijelic.de/wordpress/">Mario Bijelic</a><sup>1,3</sup>,</span>
<span class="author-block"><a href="https://www.cs.princeton.edu/~fheide/">Felix Heide</a><sup>1,3</sup></span> -->
<span class="author-block"><a href="https://ostjul.com">Julian Ost</a><sup>*</sup>,</span>
<span class="author-block"><a href="https://andrearama.github.io">Andrea Ramazzina</a><sup>*</sup>,</span>
<span class="author-block"><a href="https://amogh7joshi.github.io">Amogh Joshi</a><sup>*</sup>,</span>
<span class="author-block"><a href="">Maximilian Bömer</a><sup></sup>,</span>
<span class="author-block"><a href="http://mariobijelic.de/wordpress/">Mario Bijelic</a><sup></sup>,</span>
<span class="author-block"><a href="https://www.cs.princeton.edu/~fheide/">Felix Heide</a><sup></sup></span>
</div>
<!-- <div class="is-size-5 publication-authors">
<span class="author-block"><sup>1</sup>Princeton University,</span>
<span class="author-block"><sup>2</sup>Mercedes-Benz,</span>
<span class="author-block"><sup>3</sup>Torc Robotics</span>
</div> -->
<div class="is-size-6 publication-authors" style="margin-top: 10px;">
<span class="author-block">*indicates equal contribution</span>
</div>
<div class="column has-text-centered">
<div class="publication-links">
<!-- PDF Link. -->
<span class="link-block">
<a href="https://light.princeton.edu/wp-content/uploads/2025/08/lsd_3d.pdf"
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>
<span class="link-block">
<a href="https://arxiv.org/abs/2508.19204"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="ai ai-arxiv"></i>
</span>
<span>arXiv</span>
</a>
</span>
<!-- Video Link. -->
<!-- <span class="link-block">
<a href=""
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-youtube"></i>
</span>
<span>Video</span>
</a>
</span> -->
<!-- Code Link. -->
<span class="link-block">
<a href=""
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Code (tba)</span>
</a>
</span>
<!-- Dataset Link. -->
<!-- <span class="link-block">
<a href=""
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="far fa-images"></i>
</span>
<span>Data</span>
</a>
</span> -->
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<!-- Horizontal Scrolling Videos Section -->
<section class="video-marquee-section">
<div class="video-marquee-container">
<div class="video-marquee-track" id="videoMarqueeTrack">
<!-- Videos will be dynamically added via JavaScript -->
</div>
</div>
</section>
<section class="hero is-light is-small">
<div class="hero-body">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Interactive Generated Scenes</h2>
</div>
</div>
</div>
<div class="container">
<div class="rolling-carousel-container">
<div class="rolling-carousel-track">
<div class="carousel-slide prev-slide">
<div class="slide-preview">
<video class="preview-video" muted loop playsinline autoplay>
<source src="assets/video/high_views/93_4_highview_rgb.mp4" type="video/mp4">
</video>
<div class="preview-overlay">
<h4>Scene 3: Urban Scene</h4>
<!-- <p>Click to view</p> -->
</div>
</div>
</div>
<div class="carousel-slide active-slide">
<div class="viewer-container gpu-accelerated optimized-renderer" id="viewer-active"></div>
<div class="viewer-controls-overlay">
<ul>
<li>Click + drag: Pitch & Rotate Scene</li>
<li>Scroll: Zoom in/out</li>
<li>WASD: Move Camera</li>
<li>QERF: Rotate Scene & Up/Down</li>
</ul>
</div>
<div class="slide-info">
<h4>Scene 1: Winter Scene</h4>
<h3>An Urban Winter Street Scene.</h3>
<div class="loading-status" style="display: none;">
<div class="loading-spinner"></div>
<span class="loading-text">Loading...</span>
</div>
</div>
</div>
<div class="carousel-slide next-slide">
<div class="slide-preview">
<video class="preview-video" muted loop playsinline autoplay>
<source src="assets/video/high_views/21_3_night_highview_rgb.mp4" type="video/mp4">
</video>
<div class="preview-overlay">
<h4>Scene 2: Night Scene</h4>
<!-- <p>Click to view</p> -->
</div>
</div>
</div>
</div>
</div>
<div class="carousel-controls">
<button class="carousel-btn button is-normal is-rounded is-dark" id="prevBtn">Previous</button>
<button class="carousel-btn button is-normal is-rounded is-dark" id="nextBtn">Next</button>
</div>
<div class="dots">
<span class="dot active" onclick="currentSlide(1)"></span>
<span class="dot" onclick="currentSlide(2)"></span>
<span class="dot" onclick="currentSlide(3)"></span>
<span class="dot" onclick="currentSlide(4)"></span>
</div>
</div>
</div>
</section>
<!-- Driving Videos Grid Section -->
<section class="section">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Rendered Driving Trajectories</h2>
<div class="content has-text-justified">
<p>Below we show driving trajectories rendered from our generated 3D scenes. All videos are rendered in real-time and feature diverse geometry, lighting, and weather conditions.</p>
</div>
</div>
</div>
<div class="columns is-centered">
<div class="column is-four-fifths">
<div class="single-driving-video">
<video class="driving-video-single" autoplay muted loop playsinline preload="auto">
<source src="assets/video/4x4_driving_high_res.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
</div>
</div>
</div>
</div>
</section>
<section class="section">
<div class="container is-max-desktop">
<!-- Abstract. -->
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Summary</h2>
<div class="content has-text-justified">
<p> Large-scale scene data is essential for training and testing in robot learning. Neural reconstruction methods have promised the capability of reconstructing large physically-grounded outdoor scenes from captured sensor data. However, these methods have baked-in static environments and only allow for limited scene control -- they are functionally constrained in scene and trajectory diversity by the captures from which they are reconstructed. In contrast, generating driving data with recent image or video diffusion models offers control, however, at the cost of geometry grounding and causality. We aim to bridge this gap and present a method that directly generates large-scale 3D driving scenes with accurate geometry, allowing for causal novel view synthesis with object permanence and explicit 3D geometry estimation.</p>
</div>
</div>
</div>
<!--/ Abstract. -->
<!-- Method Overview -->
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Geometry-Grounded Large-Scale 3D Scene Generation</h2>
<div class="content has-text-justified method-description">
<p>We generate a large-scale scene as a combination of a coarse geometric layout, an environment map, and a set of Gaussians for texture details. The geometric layout is either generated, conditioned on a map, or predicted from point-cloud data and guides the overall scene structure. We can further control the setting with a scene prompt, for example, specifying time-of-day, season, and weather. Through Geometry-Grounded Distillation Sampling (GGDS), we then further optimize the Gaussian-based scene representation by leveraging 2D priors from the conditional latent diffusion model through consistent diffusion sampling (inversion) and image-space optimization -- together with a set of geometry-grounding regularizers -- and generate a causal large-scale scene representation.</p>
</div>
<div class="method-figure">
<img src="assets/images/method_overview.webp" alt="Method Overview" class="method-image">
</div>
</div>
</div>
<!--/ Method Overview -->
<!-- Results -->
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Example Scenes</h2>
<div class="content has-text-justified results-description">
<p>We visualize 3D scenes generated via our method, alongside the corresponding map of surface normals and selection of novel viewpoints at street level for each of them. In the first two columns, we provide samples of scenes with diversity in time-of-day, season, location, and scene type. In the third column, we provide examples of generated scenes with a point cloud condition and map layout condition. We confirm the method generates diverse, explicit, and causal 3D scenes.</p>
</div>
<div class="results-figure">
<img src="assets/images/main_results.webp" alt="Main Results" class="results-image">
</div>
</div>
</div>
<!--/ Results -->
<!-- Baseline Comparisons -->
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Baseline Comparisons</h2>
<div class="content has-text-justified baseline-description">
<p>Our approach generates an accurate and 3D-consistent scene representation, enabling high-quality novel view synthesis and the generation of unlimited off-trajectory viewpoints. Existing methods, such as Vista combined with Gaussian Splatting, GEN3C and MagicDrive3D, which generate driving videos, struggle with generating consistent and 3D-plausible scenes for generating novel driving trajectories.</p>
</div>
<div class="baseline-videos">
<div class="columns is-multiline">
<div class="column is-half">
<video class="baseline-video" muted autoplay loop playsinline preload="metadata">
<source src="assets/video/baseline_gen3c_u_turn.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
</div>
<div class="column is-half">
<video class="baseline-video" muted autoplay loop playsinline preload="metadata">
<source src="assets/video/baseline_vista.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
</div>
<div class="column is-half">
<video class="baseline-video" muted autoplay loop playsinline preload="metadata">
<source src="assets/video/baseline_gen3c_c_barrel_roll.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
</div>
<div class="column is-half">
<video class="baseline-video" muted autoplay loop playsinline preload="metadata">
<source src="assets/video/baseline_magicdrive3d.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
</div>
</div>
</div>
</div>
</div>
<!--/ Baseline Comparisons -->
<!-- Composability with Dynamic Actors -->
<!-- <div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Composability with Dynamic Actors</h2>
<div class="content has-text-justified composability-caption">
<p>We simulate driving trajectories in a residential street scenes for a Waymo-representative sensor stack. From bottom to top, we show a third-person view of the ego capture vehicle followed by a set of rendered front cameras. Both columns correspond to a different timestamp of the same traffic configurations + fully generated scene & asset models. The second vehicle is hidden behind the corner in the first frame.</p>
</div>
<div class="composability-images">
<div class="columns is-mobile">
<div class="column is-half">
<img src="assets/images/composability_1.png" alt="Composability Frame 1" class="composability-image">
</div>
<div class="column is-half">
<img src="assets/images/composability_2.png" alt="Composability Frame 2" class="composability-image">
</div>
</div>
</div>
</div>
</div> -->
<!--/ Composability with Dynamic Actors -->
<!-- BibTeX -->
<div class="columns is-centered">
<div class="column is-four-fifths">
<h2 class="title is-3 has-text-centered">BibTeX</h2>
<pre><code>
@article{ost2025lsd3d,
title = {LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding},
author = {Ost, Julian and Ramazzina, Andrea and Joshi, Amogh and
Bömer, Maximilian and Bijelic, Mario and Heide, Felix},
booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
volume = {40},
number = {10}
pages = {8177-8187}
year = {2026},
}
</code></pre>
</div>
</div>
<!--/ BibTeX -->
</div>
</section>
<footer class="footer">
<div class="container">
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>This website is maintained by <a href="https://github.com/princeton-computational-imaging">princeton-computational-imaging</a>.
</p>
</div>
</div>
</div>
</div>
</footer>
<script type="importmap">
{
"imports": {
"three": "https://cdnjs.cloudflare.com/ajax/libs/three.js/0.174.0/three.module.js",
"@sparkjsdev/spark": "https://sparkjs.dev/releases/spark/0.1.8/spark.module.js"
}
}
</script>
<script type="module" src="static/js/script_optimized.js"></script>
<!-- Uncomment the line below to enable performance monitoring -->
<!-- <script type="module" src="static/js/performance-monitor.js"></script> -->
</body>
</html>
WordPress