Correlation‐Aware Multiple Importance Sampling for Bidirectional Rendering Algorithms

Combining diverse sampling techniques via multiple importance (MIS) is key to achieving robustness in modern Monte Carlo light transport simulation. Many such methods additionally employ correlated path boost efficiency. Photon mapping, bidirectional tracing, and path-reuse algorithms construct sets of paths that share a common prefix. This correlation ignored by classical MIS heuristics, which can result poor technique combination noisy images. We propose practical robust solution problem. Our idea incorporate knowledge into the balance heuristic, based on known densities are already required for MIS. correlation-aware heuristic achieve considerably lower error than while avoiding computational memory overhead.