WorldScore: A Unified Evaluation Benchmark for World Generation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Duan, Haoyi, Yu, Hong-Xing, Chen, Sirui, Fei-Fei, Li, Wu, Jiajun
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917112877940736
author Duan, Haoyi
Yu, Hong-Xing
Chen, Sirui
Fei-Fei, Li
Wu, Jiajun
author_facet Duan, Haoyi
Yu, Hong-Xing
Chen, Sirui
Fei-Fei, Li
Wu, Jiajun
contents We introduce the WorldScore benchmark, the first unified benchmark for world generation. We decompose world generation into a sequence of next-scene generation tasks with explicit camera trajectory-based layout specifications, enabling unified evaluation of diverse approaches from 3D and 4D scene generation to video generation models. The WorldScore benchmark encompasses a curated dataset of 3,000 test examples that span diverse worlds: static and dynamic, indoor and outdoor, photorealistic and stylized. The WorldScore metrics evaluate generated worlds through three key aspects: controllability, quality, and dynamics. Through extensive evaluation of 19 representative models, including both open-source and closed-source ones, we reveal key insights and challenges for each category of models. Our dataset, evaluation code, and leaderboard can be found at https://haoyi-duan.github.io/WorldScore/
format Preprint
id arxiv_https___arxiv_org_abs_2504_00983
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle WorldScore: A Unified Evaluation Benchmark for World Generation
Duan, Haoyi
Yu, Hong-Xing
Chen, Sirui
Fei-Fei, Li
Wu, Jiajun
Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
We introduce the WorldScore benchmark, the first unified benchmark for world generation. We decompose world generation into a sequence of next-scene generation tasks with explicit camera trajectory-based layout specifications, enabling unified evaluation of diverse approaches from 3D and 4D scene generation to video generation models. The WorldScore benchmark encompasses a curated dataset of 3,000 test examples that span diverse worlds: static and dynamic, indoor and outdoor, photorealistic and stylized. The WorldScore metrics evaluate generated worlds through three key aspects: controllability, quality, and dynamics. Through extensive evaluation of 19 representative models, including both open-source and closed-source ones, we reveal key insights and challenges for each category of models. Our dataset, evaluation code, and leaderboard can be found at https://haoyi-duan.github.io/WorldScore/
title WorldScore: A Unified Evaluation Benchmark for World Generation
topic Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.00983