hongzhengdong / slahmr

Decoupling Human and Camera Motion from Videos in the Wild

Official PyTorch implementation of the paper Decoupling Human and Camera Motion from Videos in the Wild

Project page

ArXiv

Getting started

This code was tested on Ubuntu 22.04 LTS and requires a CUDA-capable GPU

1. Clone repository and submodules

git clone --recursive https://github.com/vye16/slahmr.git

or initialize submodules if already cloned

git submodule update --init --recursive

2. Set up conda environment

conda env create -f env.yaml
conda activate slahmr

Install current source repo

pip install -e .

Install ViTPose

pip install -v -e third_party/PHALP_plus/ViTPose

and DROID-SLAM (will take a while)

cd third_party/DROID-SLAM
python setup.py install

3. Download models from here.

gdown https://drive.google.com/uc?id=1GXAd-45GzGYNENKgQxFQ4PHrBp8wDRlW
unzip -q slahmr_dependencies.zip
rm slahmr_dependencies.zip

Data

We provide configurations for dataset formats in slahmr/confs/data:

1. Posetrack in slahmr/confs/data/posetrack.yaml
2. Egobody in slahmr/confs/data/egobody.yaml
3. 3DPW in slahmr/confs/data/3dpw.yaml
4. Custom video in slahmr/confs/data/video.yaml

Please make sure to update all paths to data in the config files.

We include tools to both process existing datasets we evaluated on in the paper, and to process custom data and videos. We include experiments from the paper on the Egobody, Posetrack, and 3DPW datasets.

If you want to run on a large number of videos, or if you want to select specific people tracks for optimization, we recommend preprocesing in advance. For a single downloaded video, there is no need to run preprocessing in advance.

From the slahmr/preproc directory, run PHALP on all your sequences

python launch_phalp.py --type <DATASET_TYPE> --root <DATASET_ROOT> --split <DATASET_SPLIT> --gpus <GPUS>

and run DROID-SLAM on all your sequences

python launch_slam.py --type <DATASET_TYPE> --root <DATASET_ROOT> --split <DATASET_SPLIT> --gpus <GPUS>

You can also update the paths to datasets in slahmr/preproc/datasets.py for repeated use.

Run the code

Make sure all checkpoints have been unpacked _DATA. We use hydra to launch experiments, and all parameters can be found in slahmr/confs/config.yaml. If you would like to update any aspect of logging or optimization tuning, update the relevant config files.

From the slahmr directory,

python run_opt.py data=<DATA_CFG> run_opt=True run_vis=True

We've provided a helper script launch.py for launching many optimization jobs in parallel. You can specify job-specific arguments with a job spec file, such as the example files in job_specs, and batch-specific arguments shared across all jobs as

python launch.py --gpus 1 2 -f job_specs/pt_val_shots.txt -s data=posetrack exp_name=posetrack_val

We've also provided a separate run_vis.py script for running visualization in bulk.

BibTeX

If you use our code in your research, please cite the following paper:

@inproceedings{ye2023slahmr,
    title={Decoupling Human and Camera Motion from Videos in the Wild},
    author={Ye, Vickie and Pavlakos, Georgios and Malik, Jitendra and Kanazawa, Angjoo},
    booktitle={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
    month={June},
    year={2023}
}