We introduce differentiable indirection -- a novel learned primitive that employs differentiable multi-scale lookup tables as an effective substitute for traditional compute and data operations across the graphics pipeline. We demonstrate its flexibility on a number of graphics tasks, i.e., geometric and image representation, texture mapping, shading, and radiance field representation. In all cases, differentiable indirection seamlessly integrates into existing architectures, trains rapidly, and yields both versatile and efficient results.

翻译：我们提出可微间接引用（differentiable indirection）——一种新颖的可学习基元，通过采用可微分多尺度查找表，有效替代图形管线中传统的计算与数据操作。我们在多项图形任务中验证其灵活性，包括几何与图像表示、纹理映射、着色以及辐射场表示。在所有场景下，可微间接引用均能无缝集成至现有架构中，实现快速训练，并生成兼具通用性与高效性的结果。