OmniHuman: A Large-scale Dataset and Benchmark for Human-Centric Video Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917550040809472 |
|---|---|
| author | Zhu, Lei Cai, Xing Chen, Yingjie Li, Yiheng Yang, Binxin Liu, Hao Chen, Jie Li, Chen LYu, Jing |
| author_facet | Zhu, Lei Cai, Xing Chen, Yingjie Li, Yiheng Yang, Binxin Liu, Hao Chen, Jie Li, Chen LYu, Jing |
| contents | Recent advancements in audio-video joint generation models have demonstrated impressive capabilities in content creation. However, generating high-fidelity human-centric videos in complex, real-world physical scenes remains a significant challenge. We identify that the root cause lies in the structural deficiencies of existing datasets across three dimensions: limited global scene and camera diversity, sparse interaction modeling (both person-person and person-object), and insufficient individual attribute alignment. To bridge these gaps, we present OmniHuman, a large-scale, multi-scene dataset designed for fine-grained human modeling. OmniHuman provides a hierarchical annotation covering video-level scenes, frame-level interactions, and individual-level attributes. To facilitate this, we develop a fully automated pipeline for high-quality data collection and multi-modal annotation. Complementary to the dataset, we establish the OmniHuman Benchmark (OHBench), a three-level evaluation system that provides a scientific diagnosis for human-centric audio-video synthesis. Crucially, OHBench introduces metrics that are highly consistent with human perception, filling the gaps in existing benchmarks by providing a comprehensive diagnosis across global scenes, relational interactions, and individual attributes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_18326 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OmniHuman: A Large-scale Dataset and Benchmark for Human-Centric Video Generation Zhu, Lei Cai, Xing Chen, Yingjie Li, Yiheng Yang, Binxin Liu, Hao Chen, Jie Li, Chen LYu, Jing Computer Vision and Pattern Recognition Recent advancements in audio-video joint generation models have demonstrated impressive capabilities in content creation. However, generating high-fidelity human-centric videos in complex, real-world physical scenes remains a significant challenge. We identify that the root cause lies in the structural deficiencies of existing datasets across three dimensions: limited global scene and camera diversity, sparse interaction modeling (both person-person and person-object), and insufficient individual attribute alignment. To bridge these gaps, we present OmniHuman, a large-scale, multi-scene dataset designed for fine-grained human modeling. OmniHuman provides a hierarchical annotation covering video-level scenes, frame-level interactions, and individual-level attributes. To facilitate this, we develop a fully automated pipeline for high-quality data collection and multi-modal annotation. Complementary to the dataset, we establish the OmniHuman Benchmark (OHBench), a three-level evaluation system that provides a scientific diagnosis for human-centric audio-video synthesis. Crucially, OHBench introduces metrics that are highly consistent with human perception, filling the gaps in existing benchmarks by providing a comprehensive diagnosis across global scenes, relational interactions, and individual attributes. |
| title | OmniHuman: A Large-scale Dataset and Benchmark for Human-Centric Video Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.18326 |