[CVPR 2026] MorphAny3D: Unleashing the Power of Structured Latent in 3D Morphing

Xiaokun Sun¹, Zeyu Cai¹, Hao Tang², Ying Tai¹, Jian Yang¹, Zhenyu Zhang^1*

¹Nanjing University    ²Peking University

^*Corresponding Author

🔨 Installation

Tested on Ubuntu 20.04, Python 3.10, NVIDIA A6000, CUDA 11.8, and PyTorch 2.4.0. Follow the steps below to set up the environment.

1. Clone the repo:

   git clone https://github.com/XiaokunSun/MorphAny3D.git
   cd MorphAny3D

2. Setup the environment:

   As MorphAny3D builds upon TRELLIS. You can find more details about the dependencies in the TRELLIS repository.

   bash ./setup.sh --new-env --basic --xformers --flash-attn --diffoctreerast --spconv --mipgaussian --kaolin --nvdiffrast
   conda install https://anaconda.org/pytorch3d/pytorch3d/0.7.8/download/linux-64/pytorch3d-0.7.8-py310_cu118_pyt240.tar.bz2 # Note: Please ensure the pytorch3d version matches your Python, CUDA and Torch versions

3. Download pretrained models:

   We do not modify the pretrained models of TRELLIS. The weights will be automatically downloaded when you run:

   TrellisImageTo3DPipeline.from_pretrained("microsoft/TRELLIS-image-large")

   Optionally, you can manually download the weights from HuggingFace and change the path in the above command to the local path.

   TrellisImageTo3DPipeline.from_pretrained("path/to/local/directory")

🕺 Inference

# 3D Morphing
python ./example_3Dmorphing.py

🪄 Application

# Disentangled 3D Morphing
python ./example_disentangled_3Dmorphing.py
# Dual-Target 3D Morphing
python ./example_dual_target_3Dmorphing.py
# 3D Style Transfer
python ./example_3Dstyle_transfer.py

💖 Acknowledgements

This code builds upon TRELLIS. We sincerely thank the authors for their great work and open-sourcing the code.

📚 Citation

If you find our work helpful for your research, please consider starring this repository ⭐ and citing our work:

@inproceedings{sun2026morphany3d,
  title={MorphAny3D: Unleashing the Power of Structured Latent in 3D Morphing},
  author={Sun, Xiaokun and Cai, Zeyu and and Tang, Hao and Tai, Ying and Yang, Jian and Zhang, Zhenyu},
  booktitle={CVPR},
  year={2026}
}