Official 🤗 Gradio Demo

MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers

Try V2 here: https://huggingface.co/spaces/Yiwen-ntu/MeshAnythingV2

For image or text to Artist-Created Mesh. We suggest first using Rodin to achieve text/image to dense mesh and then input the dense mesh to us.

MeshAnything converts any 3D representation into meshes created by human artists, i.e., Artist-Created Meshes (AMs).

Code: GitHub. Arxiv Paper: ArXiv.

❗️❗️❗️Important Notes:

  • Gradio doesn't support interactive wireframe rendering currently. For interactive mesh visualization, please use download the obj file and open it with MeshLab or https://3dviewer.net/.
  • The input mesh will be normalized to a unit bounding box. The up vector of the input mesh should be +Y for better results. Click Preprocess with Marching Cubes if the input mesh is a manually created mesh.
  • Limited by computational resources, MeshAnything is trained on meshes with fewer than 800 faces and cannot generate meshes with more than 800 faces. The shape of the input mesh should be sharp enough; otherwise, it will be challenging to represent it with only 800 faces. Thus, feed-forward image-to-3D methods may often produce bad results due to insufficient shape quality.
  • For point cloud input, please refer to our github repo GitHub.
Examples
Pages:

Try click random sampling and different Seed Value if the result is unsatisfying

If MeshAnything is helpful, please help to ⭐ the Github Repo. Thanks!

📋 License

S-Lab-1.0 LICENSE. Please refer to the LICENSE file for details.

📧 Contact

If you have any questions, feel free to open a discussion or contact us at yiwen002@e.ntu.edu.sg.