|
arxiv.org
|
<svg class="w-5 h-5 mr-2" id="primary_logo" data-name="primary logo" xmlns="http://www.w3.org/2000/svg" viewbox="69 -3 116 116"><path d="M486.149,277.877l-32.741,38.852c-1.286,1.372-2.084,3.777-1.365,5.5a4.705,4.705,0,0,0,4.4,2.914,4.191,4.191,0,0,0,3.16-1.563l40.191-42.714a4.417,4.417,0,0,0,.042-6.042Z" transform="translate(-358.165 -222.27)" fill="currentColor"></path><path d="M486.149,277.877l31.187-38.268c1.492-1.989,2.2-3.03,1.492-4.723a5.142,5.142,0,0,0-4.481-3.161h0a4.024,4.024,0,0,0-3.008,1.108L472.711,274.6a4.769,4.769,0,0,0,.015,6.53L520.512,332.2a3.913,3.913,0,0,0,3.137,1.192,4.394,4.394,0,0,0,4.027-2.818c.719-1.727-.076-3.438-1.4-5.23l-40.124-47.464" transform="translate(-358.165 -222.27)" fill="currentColor"></path><path d="M499.833,274.828,453.169,224.4s-1.713-2.08-3.524-2.124a4.607,4.607,0,0,0-4.338,2.788c-.705,1.692-.2,2.88,1.349,5.1l40.093,48.422" transform="translate(-358.165 -222.27)" fill="currentColor"></path></svg>
Tech Paper
|
Анализировать url
|
|
huggingface.co
|
<svg xmlns="http://www.w3.org/2000/svg" height="1em" style="flex:none;line-height:1" viewbox="0 0 24 24" width="1em"><script xmlns=""><title>HuggingFace</title><g fill="none" fill-rule="nonzero"><path d="M2.25 11.535c0-3.407 1.847-6.554 4.844-8.258a9.822 9.822 0 019.687 0c2.997 1.704 4.844 4.851 4.844 8.258 0 5.266-4.337 9.535-9.687 9.535S2.25 16.8 2.25 11.535z" fill="#FF9D0B"/><path d="M11.938 20.086c4.797 0 8.687-3.829 8.687-8.551 0-4.722-3.89-8.55-8.687-8.55-4.798 0-8.688 3.828-8.688 8.55 0 4.722 3.89 8.55 8.688 8.55z" fill="#FFD21E"/><path d="M11.875 15.113c2.457 0 3.25-2.156 3.25-3.263 0-.576-.393-.394-1.023-.089-.582.283-1.365.675-2.224.675-1.798 0-3.25-1.693-3.25-.586 0 1.107.79 3.263 3.25 3.263h-.003z" fill="#FF323D"/><path d="M14.76 9.21c.32.108.445.753.767.585.447-.233.707-.708.659-1.204a1.235 1.235 0 00-.879-1.059 1.262 1.262 0 00-1.33.394c-.322.384-.377.92-.14 1.36.153.283.638-.177.925-.079l-.002.003zm-5.887 0c-.32.108-.448.753-.768.585a1.226 1.226 0 01-.658-1.204c.048-.495.395-.913.878-1.059a1.262 1.262 0 011.33.394c.322.384.377.92.14 1.36-.152.283-.64-.177-.925-.079l.003.003zm1.12 5.34a2.166 2.166 0 011.325-1.106c.07-.02.144.06.219.171l.192.306c.069.1.139.175.209.175.074 0 .15-.074.223-.172l.205-.302c.08-.11.157-.188.234-.165.537.168.986.536 1.25 1.026.932-.724 1.275-1.905 1.275-2.633 0-.508-.306-.426-.81-.19l-.616.296c-.52.24-1.148.48-1.824.48-.676 0-1.302-.24-1.823-.48l-.589-.283c-.52-.248-.838-.342-.838.177 0 .703.32 1.831 1.187 2.56l.18.14z" fill="#3A3B45"/><path d="M17.812 10.366a.806.806 0 00.813-.8c0-.441-.364-.8-.813-.8a.806.806 0 00-.812.8c0 .442.364.8.812.8zm-11.624 0a.806.806 0 00.812-.8c0-.441-.364-.8-.812-.8a.806.806 0 00-.813.8c0 .442.364.8.813.8zM4.515 13.073c-.405 0-.765.162-1.017.46a1.455 1.455 0 00-.333.925 1.801 1.801 0 00-.485-.074c-.387 0-.737.146-.985.409a1.41 1.41 0 00-.2 1.722 1.302 1.302 0 00-.447.694c-.06.222-.12.69.2 1.166a1.267 1.267 0 00-.093 1.236c.238.533.81.958 1.89 1.405l.24.096c.768.3 1.473.492 1.478.494.89.243 1.808.375 2.732.394 1.465 0 2.513-.443 3.115-1.314.93-1.342.842-2.575-.274-3.763l-.151-.154c-.692-.684-1.155-1.69-1.25-1.912-.195-.655-.71-1.383-1.562-1.383-.46.007-.889.233-1.15.605-.25-.31-.495-.553-.715-.694a1.87 1.87 0 00-.993-.312zm14.97 0c.405 0 .767.162 1.017.46.216.262.333.588.333.925.158-.047.322-.071.487-.074.388 0 .738.146.985.409a1.41 1.41 0 01.2 1.722c.22.178.377.422.445.694.06.222.12.69-.2 1.166.244.37.279.836.093 1.236-.238.533-.81.958-1.889 1.405l-.239.096c-.77.3-1.475.492-1.48.494-.89.243-1.808.375-2.732.394-1.465 0-2.513-.443-3.115-1.314-.93-1.342-.842-2.575.274-3.763l.151-.154c.695-.684 1.157-1.69 1.252-1.912.195-.655.708-1.383 1.56-1.383.46.007.889.233 1.15.605.25-.31.495-.553.718-.694.244-.162.523-.265.814-.3l.176-.012z" fill="#FF9D0B"/><path d="M9.785 20.132c.688-.994.638-1.74-.305-2.667-.945-.928-1.495-2.288-1.495-2.288s-.205-.788-.672-.714c-.468.074-.81 1.25.17 1.971.977.721-.195 1.21-.573.534-.375-.677-1.405-2.416-1.94-2.751-.532-.332-.907-.148-.782.541.125.687 2.357 2.35 2.14 2.707-.218.362-.983-.42-.983-.42S2.953 14.9 2.43 15.46c-.52.558.398 1.026 1.7 1.803 1.308.778 1.41.985 1.225 1.28-.187.295-3.07-2.1-3.34-1.083-.27 1.011 2.943 1.304 2.745 2.006-.2.7-2.265-1.324-2.685-.537-.425.79 2.913 1.718 2.94 1.725 1.075.276 3.813.859 4.77-.522zm4.432 0c-.687-.994-.64-1.74.305-2.667.943-.928 1.493-2.288 1.493-2.288s.205-.788.675-.714c.465.074.807 1.25-.17 1.971-.98.721.195 1.21.57.534.377-.677 1.407-2.416 1.94-2.751.532-.332.91-.148.782.541-.125.687-2.355 2.35-2.137 2.707.215.362.98-.42.98-.42S21.05 14.9 21.57 15.46c.52.558-.395 1.026-1.7 1.803-1.308.778-1.408.985-1.225 1.28.187.295 3.07-2.1 3.34-1.083.27 1.011-2.94 1.304-2.743 2.006.2.7 2.263-1.324 2.685-.537.423.79-2.912 1.718-2.94 1.725-1.077.276-3.815.859-4.77-.522z" fill="#FFD21E"/></g><script xmlns=""/></svg>
Download Dataset
</a> <!-- <a href="https://vistorybench.github.io" target="_blank" rel="noopener noreferrer" class="inline-block bg-purple-600 hover:bg-purple-700 dark:bg-purple-500 dark:hover:bg-purple-600 text-white font-semibold py-3 px-8 rounded-lg shadow-lg hover:shadow-purple-400/50 dark:hover:shadow-purple-500/40 transition-all duration-300 transform hover:scale-105">
View Project Page
</a> --> <a href="https://github.com/vistorybench/vistorybench" target="_blank" rel="noopener noreferrer" class="inline-flex items-center bg-teal-600 hover:bg-teal-700 dark:bg-teal-500 dark:hover:bg-teal-600 text-white font-semibold py-3 px-8 rounded-lg shadow-lg hover:shadow-teal-400/50 dark:hover:shadow-teal-500/40 transition-all duration-300 transform hover:scale-105"> <svg class="w-5 h-5 mr-2" fill="currentColor" viewBox="0 0 24 24" aria-hidden="true"><path fill-rule="evenodd" d="M12 2C6.477 2 2 6.484 2 12.017c0 4.425 2.865 8.18 6.839 9.504.5.092.682-.217.682-.483 0-.237-.008-.868-.013-1.703-2.782.605-3.369-1.343-3.369-1.343-.454-1.158-1.11-1.466-1.11-1.466-.908-.62.069-.608.069-.608 1.003.07 1.531 1.032 1.531 1.032.892 1.53 2.341 1.088 2.91.832.092-.647.35-1.088.636-1.338-2.22-.253-4.555-1.113-4.555-4.951 0-1.093.39-1.988 1.029-2.688-.103-.253-.446-1.272.098-2.65 0 0 .84-.27 2.75 1.026A9.564 9.564 0 0112 6.844c.85.004 1.705.115 2.504.337 1.909-1.296 2.747-1.026 2.747-1.026.546 1.379.202 2.398.1 2.651.64.7 1.028 1.595 1.028 2.688 0 3.848-2.339 4.695-4.566 4.943.359.308.678.92.678 1.855 0 1.338-.012 2.419-.012 2.747 0 .268.18.58.688.482A10.001 10.001 0 0022 12.017C22 6.484 17.522 2 12 2z" clip-rule="evenodd" /></svg>
Code & Tools (GitHub)
</a> <a href="/leaderboard_detail" class="inline-flex items-center bg-slate-600 hover:bg-slate-700 dark:bg-slate-500 dark:hover:bg-slate-600 text-white font-semibold py-3 px-8 rounded-lg shadow-lg hover:shadow-slate-400/50 dark:hover:shadow-slate-500/40 transition-all duration-300 transform hover:scale-105"> <svg class="w-5 h-5 mr-2" fill="none" stroke="currentColor" viewBox="0 0 24 24" xmlns="http://www.w3.org/2000/svg"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 19v-6a2 2 0 00-2-2H5a2 2 0 00-2 2v6a2 2 0 002 2h2a2 2 0 002-2zm0 0V9a2 2 0 012-2h2a2 2 0 012 2v10m-6 0a2 2 0 002 2h2a2 2 0 002-2m0 0V5a2 2 0 012-2h2a2 2 0 012 2v14a2 2 0 01-2 2h-2a2 2 0 01-2-2z"></path></svg>
View Leaderboard
</a> </div> </div></section> </section> <div class="container mx-auto px-4 py-8"> <section id="leaderboard" class="mb-12 p-6 bg-white dark:bg-slate-800 shadow-lg rounded-lg"> <div class="flex flex-col sm:flex-row justify-between items-center mb-4 gap-4"> <h2 class="text-2xl font-semibold text-gray-700 dark:text-slate-200" data-lang-key="leaderboard_title">Leaderboard Rankings</h2> <div id="mode-toggle-buttons-summary" class="flex space-x-2"> <button id="lite-mode-btn-summary" class="px-4 py-2 rounded font-medium focus:outline-none focus:ring-2 focus:ring-offset-2 focus:ring-blue-500" data-lang-key="button_lite_mode">Lite Mode</button> <button id="full-mode-btn-summary" class="px-4 py-2 rounded font-medium focus:outline-none focus:ring-2 focus:ring-offset-2 focus:ring-blue-500" data-lang-key="button_full_mode">Full Mode</button> </div> </div> <div class="overflow-x-auto"> <table id="summary-leaderboard-table" class="min-w-full bg-white dark:bg-slate-800"> <thead id="summary-leaderboard-thead" class="bg-gray-200 dark:bg-slate-700 text-gray-600 dark:text-slate-300"> <!-- Headers will be rendered by script --> </thead> <tbody id="summary-leaderboard-tbody" class="text-gray-700 dark:text-slate-300"> <!-- Rows will be rendered by script --> </tbody> </table> </div> <div class="mt-6 text-center"> <a href="/leaderboard_detail" class="inline-block bg-blue-600 hover:bg-blue-700 text-white font-semibold py-2 px-6 rounded-lg shadow-md transition duration-150 ease-in-out focus:outline-none focus:ring-2 focus:ring-offset-2 focus:ring-blue-500" data-lang-key="view_more_details_button">
View More Details
</a> </div> </section> <!-- Task Definition Section --> <section class="py-16 sm:py-20 px-4"> <div class="container mx-auto"> <h2 class="text-3xl sm:text-4xl font-bold text-slate-900 dark:text-white mb-12 text-center">Task Definition for Story Visualization</h2> <div class="max-w-4xl mx-auto bg-white dark:bg-slate-800 p-6 sm:p-10 rounded-2xl shadow-xl border-gray-200 dark:border-gray-700"> <p class="text-slate-600 dark:text-slate-300 mb-6 text-lg leading-relaxed">The primary goal is to generate a sequence of images \(I_1, I_2, ..., I_m\) that faithfully and consistently visualize a given story, detailed through character information and a series of storyboard shot descriptions.</p> <div class="mb-8"> <h4 class="text-xl font-semibold text-sky-700 dark:text-sky-400 mb-3">Given Inputs:</h4> <ol class="list-decimal list-outside ml-5 space-y-4 text-slate-600 dark:text-slate-300"> <li> <strong>Character Information (\(C_1, ..., C_n\)):</strong> For each of the \(n\) characters, provide:
<ul class="list-disc list-outside ml-6 mt-2 space-y-1 text-sm bg-slate-50 dark:bg-slate-700/50 p-4 rounded-md shadow-sm"> <li>Textual appearance description (\(T_i\)).</li> <li>Corresponding reference image(s) (\(S_i\)) consistent with \(T_i\).</li> </ul> </li> <li> <strong>Storyboard Shots (\(Shot_1, ..., Shot_m\)):</strong> For each of the \(m\) shots, provide a textual description including:
<ul class="list-disc list-outside ml-6 mt-2 space-y-1 text-sm bg-slate-50 dark:bg-slate-700/50 p-4 rounded-md shadow-sm"> <li><strong>Setting Description:</strong> The environment and context of the current scene.</li> <li><strong>Plot Correspondence:</strong> The segment of the original narrative relevant to this shot.</li> <li><strong>Onstage Characters:</strong> A list of characters present and active in this shot.</li> <li><strong>Static Shot Description:</strong> Detailed depiction of characters' static actions, expressions, positions, and significant objects within the frame.</li> <li><strong>Shot Perspective Design:</strong> Cinematographic details like shot scale (e.g., close-up, wide shot), camera angle, and shot type.</li> </ul> </li> </ol> </div> <div> <h4 class="text-xl font-semibold text-sky-700 dark:text-sky-400 mb-3">Objective:</h4> <p class="text-slate-600 dark:text-slate-300">To generate an image sequence \(I_1, ..., I_m\) where each image \(I_j\) accurately and aesthetically portrays the specified \(Shot_j\). This includes:
</p> <ul class="list-disc list-outside ml-5 mt-2 space-y-1 text-slate-600 dark:text-slate-300"> <li>Faithful depiction of characters according to their references and descriptions.</li> <li>Correct representation of character actions, interactions, and expressions.</li> <li>Accurate portrayal of the scene setting and atmosphere.</li> <li>Adherence to the specified camera perspectives and shot compositions.</li> <li>Maintaining visual and stylistic consistency across the image sequence.</li> </ul> </div> </div> </div> </section> <!-- Key Dataset Features Section --> <section class="py-16 sm:py-20 px-4"> <div class="container mx-auto"> <h2 class="text-3xl sm:text-4xl font-bold text-slate-900 dark:text-white mb-12 text-center">
Dataset Core Characteristics
</h2> <div class="grid md:grid-cols-2 lg:grid-cols-4 gap-8"> <div class="feature-card bg-white dark:bg-slate-800 p-6 rounded-xl shadow-lg hover:shadow-sky-500/30 dark:hover:shadow-sky-600/30 border border-transparent hover:border-sky-300 dark:hover:border-sky-600 transition-all duration-300 transform hover:-translate-y-1"> <svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke-width="1.5" stroke="currentColor" class="w-8 h-8 mb-3 text-sky-600 dark:text-sky-400"><path stroke-linecap="round" stroke-linejoin="round" d="M12 6.042A8.967 8.967 0 006 3.75c-1.052 0-2.062.18-3 .512v14.25A8.987 8.987 0 016 18c2.305 0 4.408.867 6 2.292m0-14.25a8.966 8.966 0 016-2.292c1.052 0 2.062.18 3 .512v14.25A8.987 8.987 0 0018 18a8.967 8.967 0 00-6 2.292m0-14.25v14.25" /></svg> <h3 class="text-xl font-semibold text-sky-700 dark:text-sky-400 mb-2">Rich Diversity</h3> <p class="text-sm text-slate-600 dark:text-slate-300">Encompasses varied story types (comedy, horror, fantasy) and artistic styles (anime, 3D, realistic) for comprehensive model testing.</p> </div> <div class="feature-card bg-white dark:bg-slate-800 p-6 rounded-xl shadow-lg hover:shadow-teal-500/30 dark:hover:shadow-teal-600/30 border border-transparent hover:border-teal-300 dark:hover:border-teal-600 transition-all duration-300 transform hover:-translate-y-1"> <svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke-width="1.5" stroke="currentColor" class="w-8 h-8 mb-3 text-teal-600 dark:text-teal-400"><path stroke-linecap="round" stroke-linejoin="round" d="M3.75 12h16.5m-16.5 3.75h16.5M3.75 19.5h16.5M5.625 4.5h12.75a1.875 1.875 0 010 3.75H5.625a1.875 1.875 0 010-3.75z" /></svg> <h3 class="text-xl font-semibold text-teal-700 dark:text-teal-400 mb-2">Complex Narratives</h3> <p class="text-sm text-slate-600 dark:text-slate-300">Features stories with single/multiple protagonists, intricate plots, and detailed world-building to challenge generation accuracy and consistency.</p> </div> <div class="feature-card bg-white dark:bg-slate-800 p-6 rounded-xl shadow-lg hover:shadow-purple-500/30 dark:hover:shadow-purple-600/30 border border-transparent hover:border-purple-300 dark:hover:border-purple-600 transition-all duration-300 transform hover:-translate-y-1"> <svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke-width="1.5" stroke="currentColor" class="w-8 h-8 mb-3 text-purple-600 dark:text-purple-400"><path stroke-linecap="round" stroke-linejoin="round" d="M10.5 6a7.5 7.5 0 107.5 7.5h-7.5V6z" /><path stroke-linecap="round" stroke-linejoin="round" d="M13.5 10.5H21A7.5 7.5 0 0013.5 3v7.5z" /></svg> <h3 class="text-xl font-semibold text-purple-700 dark:text-purple-400 mb-2">Character-Centric</h3> <p class="text-sm text-slate-600 dark:text-slate-300">Focuses on maintaining character consistency across shots, providing detailed descriptions and multiple reference images per character.</p> </div> <div class="feature-card bg-white dark:bg-slate-800 p-6 rounded-xl shadow-lg hover:shadow-indigo-500/30 dark:hover:shadow-indigo-600/30 border border-transparent hover:border-indigo-300 dark:hover:border-indigo-600 transition-all duration-300 transform hover:-translate-y-1"> <svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke-width="1.5" stroke="currentColor" class="w-8 h-8 mb-3 text-indigo-600 dark:text-indigo-400"><path stroke-linecap="round" stroke-linejoin="round" d="M3.75 3v11.25A2.25 2.25 0 006 16.5h12A2.25 2.25 0 0020.25 14.25V3M3.75 21h16.5M16.5 3v11.25 M8.25 3v11.25" /></svg> <h3 class="text-xl font-semibold text-indigo-700 dark:text-indigo-400 mb-2">Significant Scale</h3> <p class="text-sm text-slate-600 dark:text-slate-300">Comprises <strong>80 story segments</strong>, <strong>344 unique characters</strong>, and <strong>509 reference images</strong>, ensuring a robust evaluation base.</p> </div> </div> </div> </section> <section id="news" class="mb-12 p-6 bg-white dark:bg-slate-800 shadow-lg rounded-lg"> <h2 class="text-2xl font-semibold text-gray-700 dark:text-slate-200 mb-3" data-lang-key="news_title">News & Updates</h2> <ul class="list-disc list-inside text-gray-600 dark:text-slate-300 space-y-2"> <li><strong class="font-semibold">[2025]</strong> 🏆 Ongoing leaderboard maintenance and evaluation of new story visualization methods.</li> <li><strong class="font-semibold">[2025.08.19]</strong> 🛠️ Major code update: Full benchmark implementation released.</li> <li><strong class="font-semibold">[2025.08.12]</strong> 📄 arXiv v3 is now available.</li> <li><strong class="font-semibold">[2025.06.25]</strong> 📄 arXiv v2 has been published.</li> <li><strong class="font-semibold">[2025.05.30]</strong> 📝 Technical report v1 released on arXiv.</li> <li><strong class="font-semibold">[2025.05.21]</strong> 🚀 Initial project launch and code release.</li> </ul> </section> <section id="about" class="mb-12 p-6 bg-white dark:bg-slate-800 shadow-lg rounded-lg"> <h2 class="text-2xl font-semibold text-gray-700 dark:text-slate-200 mb-4" data-lang-key="about_title">About</h2> <div class="mb-6"> <h3 class="text-xl font-semibold text-gray-700 dark:text-slate-200 mb-2" data-lang-key="about_citation_title">Citation</h3> <p class="text-gray-600 dark:text-slate-300 mb-2" data-lang-key="about_citation_text">If you use ViStoryBench in your research, please cite our paper:</p> <pre class="bg-gray-100 dark:bg-slate-700 p-4 rounded-md overflow-x-auto"><code class="text-sm text-gray-700 dark:text-slate-200">@article{zhuang2025vistorybench,
title={ViStoryBench: Comprehensive Benchmark Suite for Story Visualization},
author={Cailin Zhuang, Ailin Huang, Yaoqi Hu, Jingwei Wu, Wei Cheng, Jiaqi Liao, Hongyuan Wang, Xinyao Liao, Weiwei Cai, Hengyuan Xu, Xuanyang Zhang, Xianfang Zeng, Zhewei Huang, Gang Yu, Chi Zhang},
journal={arXiv preprint arxiv:2505.24862},
year={2025}
}</code></pre> </div> <div> <h3 class="text-xl font-semibold text-gray-700 dark:text-slate-200 mb-2" data-lang-key="about_license_title">License</h3> <p class="text-gray-600 dark:text-slate-300" data-lang-key="about_license_text">The benchmark code and results are licensed under the MIT License.</p> </div> </section> </div> </main> <footer class="bg-gray-100 dark:bg-slate-900 text-gray-700 dark:text-slate-300 text-center py-6 mt-12 border-t border-gray-200 dark:border-slate-700"> <p data-lang-key="footer_text">Storytelling Benchmark © 2025</p> </footer> <script src="/js/main.js" type="module"></script> </svg>
|
Анализировать url
|
