CaliTex：面向视图一致 3D 纹理生成的几何校准注意力
CaliTex: Geometry-Calibrated Attention for View-Coherent 3D Texture Generation
把注意力路由与语义部件对齐、几何条件绑定，是解决多视图扩散生成中跨视图不一致的可复用思路；做 3D 纹理/多视图生成的人可直接借鉴 Part-Aligned 与 Condition-Routed 两个注意力改造模块。
原文依据 attention ambiguity, where unstructured full attention is applied indiscriminately across tokens and modalities
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论尽管基于扩散的模型带来了重大进展，当前的 3D 纹理生成系统仍受跨视图不一致的困扰——从某个视角看很有说服力的纹理，往往在与其他视角对齐时失效。我们发现该问题源于注意力歧义：无结构的全注意力被不加区分地施加于不同 token 与模态之上，导致几何混淆以及外观—结构耦合不稳定。为此，我们提出 CaliTex，一个几何校准注意力的框架，显式地将注意力与 3D 结构对齐。它引入两个模块：部件对齐注意力（Part-Aligned Attention），在语义匹配的部件之间施加空间对齐；以及条件路由注意力（Condition-Routed Attention），通过几何条件化的通路来路由外观信息，以维持空间保真度。结合两阶段扩散 Transformer，CaliTex 使几何一致性成为网络的内在行为，而非优化的副产物。实验上，CaliTex 生成的纹理无缝且视图一致，并优于开源与商业基线。
英文原文摘要 1119 字符
Despite major advances brought by diffusion-based models, current 3D texture generation systems remain hindered by cross-view inconsistency -- textures that appear convincing from one viewpoint often fail to align across others. We find that this issue arises from attention ambiguity, where unstructured full attention is applied indiscriminately across tokens and modalities, causing geometric confusion and unstable appearance-structure coupling.To address this, we introduce CaliTex, a framework of geometry-calibrated attention that explicitly aligns attention with 3D structure.It introduces two modules: Part-Aligned Attention that enforces spatial alignment across semantically matched parts, and Condition-Routed Attention which routes appearance information through geometry-conditioned pathways to maintain spatial fidelity.Coupled with a two-stage diffusion transformer, CaliTex makes geometric coherence an inherent behavior of the network rather than a byproduct of optimization.Empirically, CaliTex produces seamless and view-consistent textures and outperforms both open-source and commercial baselines.