PhySe-RPO: Physics and Semantics Guided Relative Policy Optimization for Diffusion-Based Surgical Smoke Removal

📰 ArXiv cs.AI

PhySe-RPO is a diffusion-based surgical smoke removal framework that uses physics and semantics to optimize policy restoration

advanced Published 25 Mar 2026

Action Steps

Develop a diffusion-based restoration framework
Integrate physics and semantics guidance for policy optimization
Train the model using relative policy optimization
Evaluate the model's performance on real surgical smoke removal tasks

Who Needs to Know This

This research benefits computer vision engineers and medical professionals working on surgical smoke removal, as it provides a novel approach to improve intraoperative video quality

Key Insight

💡 Physics and semantics guidance can improve the performance of diffusion-based surgical smoke removal models