dataset_family string | repo_role string | hosts_payload_data bool | license_policy string | available_releases list |
|---|---|---|---|---|
Multimodal Annotated Naturalistic Hand Understanding (MANUS) | landing/index repository | false | per-source licenses; no single dataset-wide license | [
{
"name": "MANUS-HaGRID",
"source_datasets": [
"HaGRID",
"HaGRIDv2"
],
"configs": [
"hagrid_v1",
"hagrid_v2"
],
"splits": [
"train",
"test"
],
"license_metadata": "other",
"license_name": "hagrid-specific-cc-by-sa-4.0-variant",
"sample_coun... |
MANUS Dataset Family
Multimodal Annotated Naturalistic Hand Understanding (MANUS) is a family of source-specific multimodal hand-understanding dataset releases for naturalistic hand analysis, hand-aware image generation, 2D/3D hand understanding, depth estimation, and multi-view hand representation learning.
MANUS does not use a single dataset-wide license. Each source-specific release is governed by its own upstream-compatible license. This landing repository indexes the available MANUS releases and explains the shared schema, license matrix, citation guidance, and loading pattern. It does not host source images, hand crops, depth maps, mesh renderings, or WebDataset shards.
Available Releases
| MANUS release | Source data | Payload repo content | HF license metadata | Main constraint |
|---|---|---|---|---|
| MANUS-HaGRID | HaGRID / HaGRIDv2 | hagrid_v1 and hagrid_v2 train/test WebDataset shards |
other |
HaGRID-specific CC BY-SA 4.0 variant |
| MANUS-DexYCB | DexYCB | dexycb test WebDataset shards |
cc-by-nc-4.0 |
Non-commercial use only |
Download each subset from its own repository and comply with the corresponding license:
- MANUS-HaGRID:
QFun/MANUS-HaGRID - MANUS-DexYCB:
QFun/MANUS-DexYCB
Shared MANUS Modalities
MANUS source-specific releases are designed around synchronized multimodal and multi-view annotations where available:
- full-scene RGB image;
- full-scene depth map;
- hand-region RGB crop;
- hand-region depth crop;
- 2D bounding boxes for hand regions;
- estimated MANO-style hand mesh annotations or related mesh metadata;
- rendered mesh views from canonical viewpoints such as front, rear, left, right, top, and bottom;
- companion text prompt files where provided by a source-specific payload release;
- per-sample metadata recording source dataset, source id, split, available modalities, and license metadata.
Loading
See docs/loading_all_subsets.md for examples that load MANUS-HaGRID and MANUS-DexYCB from their source-specific repositories.
Citation
If you use any MANUS subset, please cite the MANUS / HanDrawer work and the upstream source dataset for the subset you used.
MANUS / HanDrawer
@misc{fu2025handrawer,
title = {HanDrawer: Leveraging Spatial Information to Render Realistic Hands Using a Conditional Diffusion Model in Single Stage},
author = {Qifan Fu and Xu Chen and Muhammad Asad and Shanxin Yuan and Changjae Oh and Gregory Slabaugh},
year = {2025},
eprint = {2503.02127},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
doi = {10.48550/arXiv.2503.02127},
url = {https://arxiv.org/abs/2503.02127}
}
Source Dataset Citations
- For MANUS-HaGRID, cite HaGRID / HaGRIDv2 as listed in the MANUS-HaGRID repository.
- For MANUS-DexYCB, cite DexYCB as listed in the MANUS-DexYCB repository.
License
This landing repository is an index for the MANUS dataset family. It does not grant a single license for all MANUS subsets.
Each MANUS subset is governed by the license of its own repository and its upstream source dataset. See docs/license_matrix.md for details.
Documentation and utility scripts in this landing repository are provided for release indexing and loading examples. Dataset files are not hosted here.
Intended Uses
MANUS releases are intended for research on hand gesture understanding, hand-object interaction, hand-aware image generation, 2D/3D hand reasoning, depth estimation, and multimodal representation learning.
Out-of-Scope Uses
MANUS releases should not be used for biometric identification, person recognition, surveillance, or privacy-invasive applications. Users must comply with the source-specific license and applicable laws.
- Downloads last month
- -