PhySe-RPO: Physics and Semantics Guided Relative Policy Optimization for Diffusion-Based Surgical Smoke Removal
This addresses the challenge of limited paired supervision in surgical video restoration, offering a method for enhanced surgical perception, though it is incremental as it builds on existing diffusion and reinforcement learning techniques.
The paper tackled the problem of surgical smoke degrading intraoperative video quality by proposing PhySe-RPO, a diffusion-based framework that uses physics and semantics guidance to achieve robust smoke removal, showing improved results on synthetic and real robotic surgical datasets.
Surgical smoke severely degrades intraoperative video quality, obscuring anatomical structures and limiting surgical perception. Existing learning-based desmoking approaches rely on scarce paired supervision and deterministic restoration pipelines, making it difficult to perform exploration or reinforcement-driven refinement under real surgical conditions. We propose PhySe-RPO, a diffusion restoration framework optimized through Physics- and Semantics-Guided Relative Policy Optimization. The core idea is to transform deterministic restoration into a stochastic policy, enabling trajectory-level exploration and critic-free updates via group-relative optimization. A physics-guided reward imposes illumination and color consistency, while a visual-concept semantic reward learned from CLIP-based surgical concepts promotes smoke-free and anatomically coherent restoration. Together with a reference-free perceptual constraint, PhySe-RPO produces results that are physically consistent, semantically faithful, and clinically interpretable across synthetic and real robotic surgical datasets, providing a principled route to robust diffusion-based restoration under limited paired supervision.