097-tokenhmr-advancing-human-mesh-recovery-with-a-tokenized-pose-representation

https://github.com/szu-advtech-2024/097-tokenhmr-advancing-human-mesh-recovery-with-a-tokenized-pose-representation

Scientific Fields

Repository

https://github.com/SZU-AdvTech-2024/097-TokenHMR-Advancing-Human-Mesh-Recovery-with-a-Tokenized-Pose-Representation/blob/main/

# Readme

## 
1.   `conda create -n tkhmr python=3.10`
2. pytorchcuda11.8, Pytorch2.1.0
`pip install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --index-url https://download.pytorch.org/whl/cu118`
3. TokenHMR`pip install -r requirements.txt`
4. Detectron2Demo
`pip install git+https://github.com/facebookresearch/detectron2
5. Sapienssapiensinstall.sh
6. fetch_demo_data.shhttps://huggingface.co/facebook/sapienssapiens-depth-2bTokenHMR\tokenhmr\lib\models\sapiens
7. SMPL`cp data/body_models/smpl/SMPL_NEUTRAL.pkl $HOME/.cache/phalp/3D/models/smpl/`

## Demo
```shell
python tokenhmr/demo.py \
    --img_folder demo_sample/images/ \
    --batch_size=1 \
    --full_frame \
    --checkpoint data/checkpoints/tokenhmr_model_latest.ckpt \
    --model_config data/checkpoints/model_config.yaml
```

## 
https://download.is.tue.mpg.de/download.php?domain=tokenhmr&sfile=bedlam.tar.gz bedlam
https://www.dropbox.com/scl/fo/vp8v9wxw46n63w94xxnmo/AOpIPovpwNU6ucNBamGrLg8?rlkey=lmbd7cpce009gzmc41081gesi&e=1 4DHuman


```shell
TokenHMR/
 tokenhmr/
    dataset_dir/
        training_data/                      # Training data
            dataset_tars/
                coco-train-2014-pruned/
                aic-train-vitpose/
                bedlam/
|               ...                          
           ...
        evaluation_data/                    # Evaluation data
            3DPW/
            EMDB/
            emdb.npz
            3dpw_test.npz
 ...
```
4A100
`python tokenhmr/train.py datasets=mix_all experiment=tokenhmr_release`

3DPWEMDB
https://virtualhumans.mpi-inf.mpg.de/3DPW/
https://eth-ait.github.io/emdb/

https://download.is.tue.mpg.de/download.php?domain=tokenhmr&sfile=test.tar.gz

```shell
python tokenhmr/eval.py  \
    --dataset EMDB,3DPW-TEST \
    --batch_size 32 --log_freq 50 \
    --dataset_dir tokenhmr/dataset_dir/evaluation_data \
    --checkpoint data/checkpoints/tokenhmr_model.ckpt \
    --model_config data/checkpoints/model_config.yaml```

Owner

Citation (citation.txt)

@conference{REPO097,
    author = "Dwivedi, Sai Kumar and Sun, Yu and Patel, Priyanka and Feng, Yao and Black, Michael J.",
    booktitle = "IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)",
    month = "March",
    title = "{{TokenHMR}: Advancing Human Mesh Recovery with a Tokenized Pose Representation}",
    year = "2024"
}

GitHub Events