yohanshin/WHAM
Source: https://github.com/yohanshin/wham
Structured fields
{
"archived": false,
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "CVPR 2024 method recovering world-grounded 3D human motion (SMPL) from monocular video using ViTPose keypoints and DPVO/DROID-SLAM camera motion; a path from ordinary footage to retargetable body motion.",
"source_type": "repository",
"availability": "available",
"dependencies": [
"SMPL/SMPLify registration",
"ViTPose",
"DPVO or DROID-SLAM",
"PyTorch (versions not stated on page)"
],
"next_check_at": "2026-11-16T00:00:00.000Z",
"known_limitations": [
"Output file format not specified on the README page."
],
"upstream_updated_at": "2024-04-18T19:21:17.000Z"
}Evidence claims
- [upstream_claim] README: official PyTorch implementation of WHAM (Shin, Kim, Halilaj, Black, CVPR 2024); demo.py --video; options --calib for intrinsics, --estimate_local_only to skip SLAM, --run_smplify for Temporal SMPLify refinement; uses ViTPose and DPVO/DROID-SLAM. source
- [sourced_fact] Requires SMPL and SMPLify registration to fetch body models and demo data; repository lists MIT but the page does not describe SMPL/AMASS/dataset licensing and README TODO items (data preprocessing, training, Colab, custom-video demo) have unclear completion state. source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | MIT | permitted | permitted | permitted | unknown | Repository lists MIT. |
| other | unknown | restricted | prohibited | restricted | restricted | Pipeline depends on SMPL body models whose license permits only non-commercial scientific research, education and artistic projects and prohibits redistribution (see smpl-body-model-license). MIT on the WHAM code does not lift this. |
| dataset | unknown | unknown | unknown | unknown | unknown | AMASS and other training/eval datasets have their own terms; not read. |