Hi, and welcome to the codebase for our CVPR22 submission.
Feel free to explore one of our pre-trained models using one of the notebooks in the "notebooks" folder.
For example re-run the evaluation on H3.6M in the "RES H3.6M Deterministic Evaluation.ipynb" notebook.
Or train your own model via
PYTHONPATH=. python -O h36m/train.py --device cpu --config ./config/h36m.yaml
Model configuration is passed via the YAML file located in the "config" folder.
Please download the processed H3.6M dataset from here and place it in
./data/processed/h3.6m.npz
Install the requirements
pip install -r requirements.txt
If your device supports GPU acceleration please uncomment line 5 in requirements.txt.
To visualize AMASS output please additionally install
git+https://github.com/nghorbani/configer@5f2b82e5
git+https://github.com/nghorbani/human_body_prior@1936f38
git+https://github.com/nghorbani/amass@e5f0889
The main results can be found here and here.
To download and preprocess the H36M dataset automatically run
PYTHONPATH=. python ./preprocess/h36m.py
Download the AMASS datasets you want to use from https://amass.is.tue.mpg.de
To preprocess run
PYTHONPATH=. python ./preprocess/amass.py --path [path to amass datasets] --out ./data/processed/amass --datasets [List of datasets to process]
In the paper we eventually also follow the amass evaluation procedure of Aksan et al.. For the future we suggest to use their evaluation data/split as the distribution difference between train and test data is more reasonable.
@article{salzmann2022motron,
title={Motron: Multimodal Probabilistic Human Motion Forecasting},
author={Salzmann, Tim and Pavone, Marco and Ryll, Markus},
journal={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2022}
}
If you use data from AMASS or H36M, please cite the original papers as detailed on their website.