Modular Primitives for High-Performance Differentiable Rendering

TL;DR

This paper introduces a modular differentiable renderer that leverages hardware graphics pipelines for high performance, enabling efficient inverse rendering applications like facial performance capture within automatic differentiation frameworks.

Contribution

It presents a novel modular design for differentiable rendering that integrates with existing graphics hardware and frameworks, improving performance and flexibility.

Findings

Achieves high-resolution rendering with all key graphics operations.

Enables efficient inverse rendering for facial capture.

Demonstrates excellent geometric correspondence in results.

Abstract

We present a modular differentiable renderer design that yields performance superior to previous methods by leveraging existing, highly optimized hardware graphics pipelines. Our design supports all crucial operations in a modern graphics pipeline: rasterizing large numbers of triangles, attribute interpolation, filtered texture lookups, as well as user-programmable shading and geometry processing, all in high resolutions. Our modular primitives allow custom, high-performance graphics pipelines to be built directly within automatic differentiation frameworks such as PyTorch or TensorFlow. As a motivating application, we formulate facial performance capture as an inverse rendering problem and show that it can be solved efficiently using our tools. Our results indicate that this simple and straightforward approach achieves excellent geometric correspondence between rendered results and reference imagery.