Hugging Face's logo

walterzhu
/
MotionBERT

3D Human Pose Estimation
Skeleton-based Action Recognition
Mesh Recovery
Model card Files Community
2
MotionBERT / docs /pose3d.md
walterzhu's picture
walterzhu
Upload 58 files
bbde80b
preview code
|
raw
history blame contribute delete
1.22 kB

3D Human Pose Estimation

Data

  1. Download the finetuned Stacked Hourglass detections and our preprocessed H3.6M data (.pkl) here and put it to data/motion3d.

Note that the preprocessed data is only intended for reproducing our results more easily. If you want to use the dataset, please register to the Human3.6m website and download the dataset in its original format. Please refer to LCN for how we prepare the H3.6M data.

  1. Slice the motion clips (len=243, stride=81)

    python tools/convert_h36m.py

Running

Train from scratch:

python train.py \
--config configs/pose3d/MB_train_h36m.yaml \
--checkpoint checkpoint/pose3d/MB_train_h36m

Finetune from pretrained MotionBERT:

python train.py \
--config configs/pose3d/MB_ft_h36m.yaml \
--pretrained checkpoint/pretrain/MB_release \
--checkpoint checkpoint/pose3d/FT_MB_release_MB_ft_h36m

Evaluate:

python train.py \
--config configs/pose3d/MB_train_h36m.yaml \
--evaluate checkpoint/pose3d/MB_train_h36m/best_epoch.bin