UniLight：一种统一的光照表示
UniLight: A Unified Representation for Lighting
把光照（环境图、辐照度、文本、图像）统一到一个共享潜空间，可直接作为扩散图像合成的条件信号，实现跨模态光照检索与可控重打光；其对比对齐加球谐预测的辅助任务思路也可迁移到其他物理属性（材质、相机参数）的跨模态表示学习。
原文依据 a joint latent space as lighting representation, that unifies multiple modalities within a shared embedding
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论光照对视觉外观有强烈影响，然而在图像中理解和表示光照仍然极为困难。现有的光照表示多种多样，例如环境图、辐照度、球谐函数或文本，但它们彼此不兼容，限制了跨模态迁移。为此，我们提出 UniLight，一种作为光照表示的联合潜空间，在共享嵌入中统一多种模态。针对文本、图像、辐照度和环境图的模态专属编码器通过对比方式训练以对齐其表示，并辅以一个球谐函数预测任务来强化方向性理解。我们的多模态数据流水线支持大规模训练与评测，覆盖三项任务：基于光照的检索、环境图生成，以及基于扩散的图像合成中的光照控制。实验表明，我们的表示能够捕捉一致且可迁移的光照特征，从而实现跨模态的灵活操控。
英文原文摘要 1024 字符
Lighting has a strong influence on visual appearance, yet understanding and representing lighting in images remains notoriously difficult. Various lighting representations exist, such as environment maps, irradiance, spherical harmonics, or text, but they are incompatible, which limits cross-modal transfer. We thus propose UniLight, a joint latent space as lighting representation, that unifies multiple modalities within a shared embedding. Modality-specific encoders for text, images, irradiance, and environment maps are trained contrastively to align their representations, with an auxiliary spherical-harmonics prediction task reinforcing directional understanding. Our multi-modal data pipeline enables large-scale training and evaluation across three tasks: lighting-based retrieval, environment-map generation, and lighting control in diffusion-based image synthesis. Experiments show that our representation captures consistent and transferable lighting features, enabling flexible manipulation across modalities.