Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912088070291456 |
|---|---|
| author | Khirodkar, Rawal Song, Jyun-Ting Cao, Jinkun Luo, Zhengyi Kitani, Kris |
| author_facet | Khirodkar, Rawal Song, Jyun-Ting Cao, Jinkun Luo, Zhengyi Kitani, Kris |
| contents | Understanding how humans interact with each other is key to building realistic multi-human virtual reality systems. This area remains relatively unexplored due to the lack of large-scale datasets. Recent datasets focusing on this issue mainly consist of activities captured entirely in controlled indoor environments with choreographed actions, significantly affecting their diversity. To address this, we introduce Harmony4D, a multi-view video dataset for human-human interaction featuring in-the-wild activities such as wrestling, dancing, MMA, and more. We use a flexible multi-view capture system to record these dynamic activities and provide annotations for human detection, tracking, 2D/3D pose estimation, and mesh recovery for closely interacting subjects. We propose a novel markerless algorithm to track 3D human poses in severe occlusion and close interaction to obtain our annotations with minimal manual intervention. Harmony4D consists of 1.66 million images and 3.32 million human instances from more than 20 synchronized cameras with 208 video sequences spanning diverse environments and 24 unique subjects. We rigorously evaluate existing state-of-the-art methods for mesh recovery and highlight their significant limitations in modeling close interaction scenarios. Additionally, we fine-tune a pre-trained HMR2.0 model on Harmony4D and demonstrate an improved performance of 54.8% PVE in scenes with severe occlusion and contact. Code and data are available at https://jyuntins.github.io/harmony4d/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_20294 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions Khirodkar, Rawal Song, Jyun-Ting Cao, Jinkun Luo, Zhengyi Kitani, Kris Computer Vision and Pattern Recognition Understanding how humans interact with each other is key to building realistic multi-human virtual reality systems. This area remains relatively unexplored due to the lack of large-scale datasets. Recent datasets focusing on this issue mainly consist of activities captured entirely in controlled indoor environments with choreographed actions, significantly affecting their diversity. To address this, we introduce Harmony4D, a multi-view video dataset for human-human interaction featuring in-the-wild activities such as wrestling, dancing, MMA, and more. We use a flexible multi-view capture system to record these dynamic activities and provide annotations for human detection, tracking, 2D/3D pose estimation, and mesh recovery for closely interacting subjects. We propose a novel markerless algorithm to track 3D human poses in severe occlusion and close interaction to obtain our annotations with minimal manual intervention. Harmony4D consists of 1.66 million images and 3.32 million human instances from more than 20 synchronized cameras with 208 video sequences spanning diverse environments and 24 unique subjects. We rigorously evaluate existing state-of-the-art methods for mesh recovery and highlight their significant limitations in modeling close interaction scenarios. Additionally, we fine-tune a pre-trained HMR2.0 model on Harmony4D and demonstrate an improved performance of 54.8% PVE in scenes with severe occlusion and contact. Code and data are available at https://jyuntins.github.io/harmony4d/. |
| title | Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.20294 |