VBench: Comprehensive Benchmark Suite for Video Generative Models
Ziqi Huang, Yinan He, Jiashuo Yu, Fan Zhang, Chenyang Si, Yuming Jiang, Yuanhan Zhang, Tianxing Wu
Abstract
Video generation has witnessed significant advancements yet evaluating these models remains a challenge. A comprehensive evaluation benchmark for video generation is indispensable for two reasons: 1) Existing metrics do not fully align with human perceptions; 2) An ideal evaluation system should provide insights to inform future developments of video generation. To this end we present VBench a comprehensive benchmark suite that dissects "video generation quality" into specific hierarchical and disentangled dimensions each with tailored prompts and evaluation methods. VBench has three appealing properties: 1) Comprehensive Dimensions: VBench comprises 16 dimensions in video generation (e.g. subject identity inconsistency motion smoothness temporal flickering and spatial relationship etc). The evaluation metrics with fine-grained levels reveal individual models' strengths and weaknesses. 2) Human Alignment: We also provide a dataset of human preference annotations to validate our benchmarks' alignment with human perception for each evaluation dimension respectively. 3) Valuable Insights: We look into current models' ability across various evaluation dimensions and various content types. We also investigate the gaps between video and image generation models. We will open-source VBench including all prompts evaluation methods generated videos and human preference annotations and also include more video generation models in VBench to drive forward the field of video generation.
BibTeX
@inproceedings{cvpr2024_vbenchcomprehens,
title = {VBench: Comprehensive Benchmark Suite for Video Generative Models},
author = {Ziqi Huang and Yinan He and Jiashuo Yu and Fan Zhang and Chenyang Si and Yuming Jiang and Yuanhan Zhang and Tianxing Wu and Qingyang Jin and Nattapol Chanpaisit and Yaohui Wang and Xinyuan Chen and Limin Wang and Dahua Lin and Yu Qiao and Ziwei Liu},
booktitle = {CVPR 2024},
year = {2024}
}