Abstract
Laparoscopic surgery offers minimally invasive procedures with better patient outcomes, but smoke presence challenges visibilityand safety. Existing learning-based methods demand large datasets andhigh computational resources. We propose the Progressive Frequency-Aware Network (PFAN), a lightweight GAN framework for laparoscopicimage desmoking, combining the strengths of CNN and Transformerfor progressive information extraction in the frequency domain. PFAN features CNN-based Multi-scale Bottleneck-Inverting (MBI) Blocks forcapturing local high-frequency information and Locally-Enhanced Axial Attention Transformers (LAT) for efficiently handling global low-frequency information. PFAN efficiently desmokes laparoscopic imageseven with limited training data. Our method outperforms state-of-the-art approaches in PSNR, SSIM, CIEDE2000, and visual quality on the Cholec80 dataset and retains only 629K parameters. Our code and models are made publicly available at : https://github.com/jlzcode/PFAN.
Jiale Zhang (张嘉乐)
My research interests include Human-Computer Interaction, Artificial Intelligence, and Machine Learning.