HuMMan is a multi-modal 4D human dataset presented at ECVV 2022. The dataset is hosted by author caizhongang on HuggingFace and was updated on October 7, 2024. It includes subsets for motion generation (HuMMan-MoGen), 3D vision (HuMMan-Point), and depth maps (HuMMan-Recon).
Use Cases
- Training models for 4D human motion generation based on the HuMMan-MoGen subset.
- Developing 3D human reconstruction algorithms based on the HuMMan-Point subset.
- Researching multi-modal sensing and modeling of human activities based on the dataset's 4D capture.
Strengths
- Dataset is associated with a peer-reviewed ECCV 2022 oral presentation.
- Includes multiple specialized subsets (MoGen, Point, Recon) for different research tasks.
- Actively maintained, with a major update to HuggingFace in October 2024.
Limitations
- Description metadata is limited; actual data quality, scale, and column definitions require manual inspection after download.
- Row count, file formats, and license information are unknown, which may limit suitability assessment.
Provenance
- Source
- caizhongang on HuggingFace.
- Collection Method
- Likely captured via multi-modal sensors for 4D human data.
- Time Range
- null
- Freshness
- Last updated 2024-10-07 11:37:01.
- Geography
- null