NRRS: Neural Russian Roulette and Splitting

We propose a novel framework for Russian Roulette and Splitting (RRS) tailored to wavefront path tracing, a highly parallel rendering architecture that processes path states in batched, stage-wise execution for efficient GPU utilization. Traditional RRS methods, with unpredictable path counts, are fundamentally incompatible with wavefront's preallocated memory and scheduling requirements. To resolve this, we introduce a normalized RRS formulation with a bounded path count, enabling stable and memory-efficient execution. Furthermore, we pioneer the use of neural networks to learn RRS factors, presenting two models: NRRS and AID-NRRS. Both feature a carefully designed RRSNet that explicitly incorporates RRS normalization. We also introduce Mix-Depth, a path-depth-aware mechanism that adaptively regulates neural evaluation. Extensive experiments demonstrate that our method outperforms traditional heuristics and recent RRS techniques across a variety of complex scenes.