📄 ClothHMR: 3D Mesh Recovery of Humans in Diverse Clothing from Single Image

Yunqi Gao, Leyuan Liu, Yuhan Li, Changxin Gao, Yuanyuan Liu, Jingying Chen
ICMR 2025
[arXiv] | [Project Page]

📊 Results

---

🛠 Installation

git clone https://github.com/leyuanCv/ClothHMR.git
cd your_repo
conda create -n clothhmr python=3.10
conda activate clothhmr
pip install -r requirements.txt

---

🚀 Usage

Step 1: Download Pretrained Models

Download the pretrained model from the following links and place it in the data folder:
Google Drive

Step 2:

Create the images path and place your input images there.
Use sapiens to obtain the human depth map and 133 keypoints.
Save the depth maps in the images_depth folder and the pose keypoints in the images_pose folder, following the structure shown in the examples directory.

Step 3: Run Demo

# Run demo
cd apps
python  infer.py --in_dir ./examples/images --out_dir ./examples/output
#You can use TokenHMR, PyMAF, or SPIN to obtain the initial SMPL model.

---

🙏 Acknowledgements

We thank the authors of the following open-source projects for their valuable code and resources:

• ECON
• TokenHMR
• PyMAF
• SPIN
• SAPIENS

---

📖 Citation

If you find our work helpful, please consider citing our paper:

@inproceedings{Clothhmr,
  title     = {ClothHMR: 3D Mesh Recovery of Humans in Diverse Clothing from Single Image},
  author    = {Yunqi Gao, Leyuan Liu, Yuhan Li, Changxin Gao, Yuanyuan Liu, Jingying Chen},
  booktitle = {ICMR},
  year      = {2025}
}

---

📬 Contact

For any questions, feel free to contact us via [gaoyunqi@mails.ccnu.edu.cn] or open an issue in this repository.