🤖 AI Summary
This paper presents a systematic survey of recent advances in physics-based differentiable rendering for inverse rendering. It addresses three core challenges: inaccurate gradient computation with respect to scene parameters, high variance in Monte Carlo estimators, and low computational efficiency. Methodologically, the work unifies three key innovations—general theoretical modeling, unbiased gradient estimation, and GPU-native acceleration architectures—into a coherent classification framework. Specific techniques covered include differentiable path tracing, adjoint light transport, reparameterized sampling, radiometric gradient propagation, and neural-hardware co-optimization. The survey clarifies the technical evolution, identifies fundamental limitations—including gradient breakdown under discontinuous geometry or materials—and highlights practical bottlenecks such as memory overhead versus variance trade-offs. Collectively, it provides a clear roadmap for both theoretical advancement and real-world deployment of differentiable rendering.
📝 Abstract
Physics-based differentiable rendering has emerged as a powerful technique in computer graphics and vision, with a broad range of applications in solving inverse rendering tasks. At its core, differentiable rendering enables the computation of gradients with respect to scene parameters, allowing optimization-based approaches to solve various problems. Over the past few years, significant advancements have been made in both the underlying theory and the practical implementations of differentiable rendering algorithms. In this report, we provide a comprehensive overview of the current state of the art in physics-based differentiable rendering, focusing on recent advances in general differentiable rendering theory, Monte Carlo sampling strategy, and computational efficiency.