Prevailing 3D texture generation methods, which often rely on multi-view fusion, are frequently hindered by inter-view inconsistencies and incomplete coverage of complex surfaces, limiting the fidelity and completeness of the generated content. To overcome these challenges, we introduce TEXTRIX, a native 3D attribute generation framework for high-fidelity texture synthesis and downstream applications such as precise 3D part segmentation. Our approach constructs a latent 3D attribute grid and leverages a Diffusion Transformer equipped with sparse attention, enabling direct coloring of 3D models in volumetric space and fundamentally avoiding the limitations of multi-view fusion. Built upon this native representation, the framework naturally extends to high-precision 3D segmentation by training the same architecture to predict semantic attributes on the grid. Extensive experiments demonstrate state-of-the-art performance on both tasks, producing seamless, high-fidelity textures and accurate 3D part segmentation with precise boundaries.
@article{arxiv.2512.02993,
title = {TEXTRIX: Latent Attribute Grid for Native Texture Generation and Beyond},
author = {Yifei Zeng and Yajie Bao and Jiachen Qian and Shuang Wu and Youtian Lin and Hao Zhu and Buyu Li and Feihu Zhang and Xun Cao and Yao Yao},
journal= {arXiv preprint arXiv:2512.02993},
year = {2025}
}