conda create -n adhmr python=3.8 -y
conda activate adhmr
conda install pytorch==1.12.0 torchvision==0.13.0 torchaudio==0.12.0 cudatoolkit=11.3 -c pytorch -y
pip install mmcv-full==1.7.1 -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.12.0/index.html
pip install -r requirements.txt
# install pytorch3d
pip install "git+https://github.com/facebookresearch/pytorch3d.git"cd HMR-Scorer/- download all datasets
- process all datasets into HumanData format, except the following:
- Human3.6M.
- follow OSX in preparing this dataset.
- follow OSX in preparing pretrained ViTPose models. Download the ViTPose pretrained weights for ViT-small and ViT-huge from here.
- download SMPL-X and SMPL body models.
- download HMR-Scorer test datasets (DNA-Rendering and GTA-Human).
The file structure should be like:
HMR-Scorer/
├── common/
│ └── utils/
│ └── human_model_files/ # body model
│ ├── smpl/
│ │ ├──SMPL_NEUTRAL.pkl
│ │ ├──SMPL_MALE.pkl
│ │ └──SMPL_FEMALE.pkl
│ └── smplx/
│ ├──MANO_SMPLX_vertex_ids.pkl
│ ├──SMPL-X__FLAME_vertex_ids.npy
│ ├──SMPLX_NEUTRAL.pkl
│ ├──SMPLX_to_J14.pkl
│ ├──SMPLX_NEUTRAL.npz
│ ├──SMPLX_MALE.npz
│ └──SMPLX_FEMALE.npz
├── data/
├── main/
├── output/
├── pretrained_models/ # pretrained ViT-Pose, SMPLer_X and mmdet models
│ ├── smpler_x_s32.pth.tar
│ ├── smpler_x_b32.pth.tar
│ ├── smpler_x_l32.pth.tar
│ ├── smpler_x_h32.pth.tar
│ ├── vitpose_small.pth
│ ├── vitpose_base.pth
│ ├── vitpose_large.pth
│ └── vitpose_huge.pth
└── dataset/
├── 3DPW/
├── Human36M/
├── HI4D
├── BEDLAM/
├── RenBody/
├── GTA_Human2/
├── CHI3D/
├── InstaVariety/
├── SPEC/
├── cache_scorer_eval/ # HMR-Scorer test datasets
└── preprocessed_datasets/ # HumanData files
cd ADHMR/- follow ScoreHypo to prepare data.
- download ckeckpoint of HMR-Scorer from OneDrive
- put it under
HMR-Scorer/output/
- put it under
- download checkpoint of ADHMR from OneDrive
- put them under
ADHMR/experiment/hyponet
- put them under
# To eval the model HMR-Scorer/output/{TRAIN_OUTPUT_DIR}
export PYTHONPATH=$PYTHONPATH:/path_to/ADHMR/HMR-Scorer
cd ./HMR-Scorer/main/
JOB_NAME=GTA_Human2 / RenBody_HiRes
torchrun test_scorer.py --num_gpus 1 --exp_name output/scorer_test_${JOB_NAME} --result_path train_scorer_b5_2d_1118_all_loss_20241120_144943 --ckpt_idx 20 --testset ${JOB_NAME}
- NUM_GPU = 1 is recommended for testing
# To eval on 3DPW
torchrun --nproc_per_node=2 --master_port=23452 main/main.py --config config/test/test-3dpw-custom.yaml --exp experiment/scorenet --doc 3dpw --validate --multihypo_n 100 --batch_size 80
# To eval on Human3.6M
torchrun --nproc_per_node=2 --master_port=23452 main/main.py --config config/test/test-h36m-custom.yaml --exp experiment/scorenet --doc h36m --validate --multihypo_n 100 --batch_size 80
- NUM_GPU = 1 is recommended for testing
If you find our work useful for your research, please consider citing the paper:
@inproceedings{
title={ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization},
author={},
booktitle={ICML},
year={2025}
}
Distributed under the S-Lab License. See LICENSE for more information.
This repo is built on the excellent work SMPLer-X, ScoreHypo. Thanks for their greate projects.