SUPRA-Tex: 3D Generation with Super-Aligned Texture
Update README authors and initial release notes
chensjtuf70d5057ae2e
diff --git a/README.md b/README.mdindex 81080f5..09baa2f 100644--- a/README.md+++ b/README.md@@ -1,17 +1,33 @@<h1 align="center">SUPRA-Tex: 3D Generation with Super-Aligned Texture</h1>-<p align="center">- Chen Yang · Jiemin Fang · Fanpeng Meng · Zihao Huang<br>- Jichen Pang · Guanjun Wu · Zhikuan Bao · Qi Tian-</p>-<p align="center"><a href="https://supra-3d.github.io/supratex/"><img src="https://img.shields.io/badge/Project-Page-blue" alt="Project page"></a>+ <img src="https://img.shields.io/badge/Paper-arXiv-b31b1b" alt="arXiv paper" title="arXiv link coming soon"><a href="https://huggingface.co/spaces/UniLat3D/Supra-Tex"><img src="https://img.shields.io/badge/Demo-Hugging_Face-blueviolet" alt="Online demo"></a><a href="https://huggingface.co/supra3d/supra-tex"><img src="https://img.shields.io/badge/Models-Hugging_Face-yellow" alt="Model weights"></a><a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-green" alt="Source license"></a></p>+<p align="center">+ <a href="https://chensjtu.github.io/">Chen Yang</a><sup>1*</sup>,+ <a href="https://jaminfong.cn/">Jiemin Fang</a><sup>1*†✉</sup>,+ <a href="https://mfp0610.github.io/">Fanpeng Meng</a><sup>1*</sup>,+ <a href="https://inso-13.github.io/">Zihao Huang</a><sup>1</sup>,<br>+ <a href="https://supra-3d.github.io/supratex/#paper">Jichen Pang</a><sup>2</sup>,+ <a href="https://wuguanjun0627.github.io/">Guanjun Wu</a><sup>3</sup>,+ <a href="https://supra-3d.github.io/supratex/#paper">Zhikuan Bao</a><sup>1</sup>,+ <a href="https://www.qitian1987.com/index.html">Qi Tian</a><sup>1✉</sup>+</p>++<p align="center">+ <sup>1</sup>Huawei Inc. + <sup>2</sup>Shanghai Jiao Tong University + <sup>3</sup>Huazhong University of Science and Technology<br>+ <sup>*</sup>Equal contribution + <sup>†</sup>Project lead + <sup>✉</sup>Corresponding authors+</p>+https://github.com/user-attachments/assets/208ede7e-6db7-4f23-8deb-61d2ac5541f5**SUPRA-Tex** generates textured 3D assets from a single image, preserving details@@ -30,6 +46,10 @@ a relightable GLB with base color, metallic, and roughness textures.Learn more in the [method overview](docs/method.md), or try the[online demo](https://huggingface.co/spaces/UniLat3D/Supra-Tex).+## Release Notes++- **[2026-10-16]** Initial release of the [project page](https://supra-3d.github.io/supratex/) and README. Inference code will follow.+## InstallationUse **Linux** with an **NVIDIA GPU** and **CUDA 12.4**. The setup below creates a@@ -71,8 +91,9 @@ hf auth login```The inference command below downloads the required models on first use and-reuses them on later runs. Both inference entrypoints use the release's-[`pipeline.json`](pipeline.json) for model configuration.+reuses them on later runs. Both inference entrypoints read model configuration+from the selected model release; source defaults are in+[`configs/pipeline.json`](configs/pipeline.json).See [model weights](docs/weights.md) for the complete dependency list and licenses.## Usage@@ -115,15 +136,6 @@ This entrypoint loads the completion model and TRELLIS.2 VAEs. The[material frontend](docs/frontend.md) can prepare the reference-view material maps.For an additional back-view observation, see [dual-view texturing](docs/dual_view.md).-## Training--We provide training code for the 3D completion model and the Qwen material adapter.--- **3D completion:** Follow the [training guide](docs/training.md) to prepare- aligned shape, material, and partial-observation latents and configure a run.-- **Material adapter:** Follow the [material training guide](docs/training.md#material-adapter-training)- to encode paired-image data and launch the Qwen-Image-2.1 adapter trainer.-## AcknowledgmentsOur implementation builds on [TRELLIS.2](https://github.com/microsoft/TRELLIS.2)