Evaluation of Text-to-Video Generation Models: A Dynamics Perspective
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914853295226880 |
|---|---|
| author | Liao, Mingxiang Lu, Hannan Zhang, Xinyu Wan, Fang Wang, Tianyu Zhao, Yuzhong Zuo, Wangmeng Ye, Qixiang Wang, Jingdong |
| author_facet | Liao, Mingxiang Lu, Hannan Zhang, Xinyu Wan, Fang Wang, Tianyu Zhao, Yuzhong Zuo, Wangmeng Ye, Qixiang Wang, Jingdong |
| contents | Comprehensive and constructive evaluation protocols play an important role in the development of sophisticated text-to-video (T2V) generation models. Existing evaluation protocols primarily focus on temporal consistency and content continuity, yet largely ignore the dynamics of video content. Dynamics are an essential dimension for measuring the visual vividness and the honesty of video content to text prompts. In this study, we propose an effective evaluation protocol, termed DEVIL, which centers on the dynamics dimension to evaluate T2V models. For this purpose, we establish a new benchmark comprising text prompts that fully reflect multiple dynamics grades, and define a set of dynamics scores corresponding to various temporal granularities to comprehensively evaluate the dynamics of each generated video. Based on the new benchmark and the dynamics scores, we assess T2V models with the design of three metrics: dynamics range, dynamics controllability, and dynamics-based quality. Experiments show that DEVIL achieves a Pearson correlation exceeding 90% with human ratings, demonstrating its potential to advance T2V generation models. Code is available at https://github.com/MingXiangL/DEVIL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_01094 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluation of Text-to-Video Generation Models: A Dynamics Perspective Liao, Mingxiang Lu, Hannan Zhang, Xinyu Wan, Fang Wang, Tianyu Zhao, Yuzhong Zuo, Wangmeng Ye, Qixiang Wang, Jingdong Computer Vision and Pattern Recognition Comprehensive and constructive evaluation protocols play an important role in the development of sophisticated text-to-video (T2V) generation models. Existing evaluation protocols primarily focus on temporal consistency and content continuity, yet largely ignore the dynamics of video content. Dynamics are an essential dimension for measuring the visual vividness and the honesty of video content to text prompts. In this study, we propose an effective evaluation protocol, termed DEVIL, which centers on the dynamics dimension to evaluate T2V models. For this purpose, we establish a new benchmark comprising text prompts that fully reflect multiple dynamics grades, and define a set of dynamics scores corresponding to various temporal granularities to comprehensively evaluate the dynamics of each generated video. Based on the new benchmark and the dynamics scores, we assess T2V models with the design of three metrics: dynamics range, dynamics controllability, and dynamics-based quality. Experiments show that DEVIL achieves a Pearson correlation exceeding 90% with human ratings, demonstrating its potential to advance T2V generation models. Code is available at https://github.com/MingXiangL/DEVIL. |
| title | Evaluation of Text-to-Video Generation Models: A Dynamics Perspective |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.01094 |