Investigating the Impact of Various Loss Functions and Learnable Wiener Filter for Laparoscopic Image Desmoking

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Chengyu, Liu, Chengjun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908534931718144
author Yang, Chengyu
Liu, Chengjun
author_facet Yang, Chengyu
Liu, Chengjun
contents To rigorously assess the effectiveness and necessity of individual components within the recently proposed ULW framework for laparoscopic image desmoking, this paper presents a comprehensive ablation study. The ULW approach combines a U-Net based backbone with a compound loss function that comprises mean squared error (MSE), structural similarity index (SSIM) loss, and perceptual loss. The framework also incorporates a differentiable, learnable Wiener filter module. In this study, each component is systematically ablated to evaluate its specific contribution to the overall performance of the whole framework. The analysis includes: (1) removal of the learnable Wiener filter, (2) selective use of individual loss terms from the composite loss function. All variants are benchmarked on a publicly available paired laparoscopic images dataset using quantitative metrics (SSIM, PSNR, MSE and CIEDE-2000) alongside qualitative visual comparisons.
format Preprint
id arxiv_https___arxiv_org_abs_2509_09849
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Investigating the Impact of Various Loss Functions and Learnable Wiener Filter for Laparoscopic Image Desmoking
Yang, Chengyu
Liu, Chengjun
Computer Vision and Pattern Recognition
To rigorously assess the effectiveness and necessity of individual components within the recently proposed ULW framework for laparoscopic image desmoking, this paper presents a comprehensive ablation study. The ULW approach combines a U-Net based backbone with a compound loss function that comprises mean squared error (MSE), structural similarity index (SSIM) loss, and perceptual loss. The framework also incorporates a differentiable, learnable Wiener filter module. In this study, each component is systematically ablated to evaluate its specific contribution to the overall performance of the whole framework. The analysis includes: (1) removal of the learnable Wiener filter, (2) selective use of individual loss terms from the composite loss function. All variants are benchmarked on a publicly available paired laparoscopic images dataset using quantitative metrics (SSIM, PSNR, MSE and CIEDE-2000) alongside qualitative visual comparisons.
title Investigating the Impact of Various Loss Functions and Learnable Wiener Filter for Laparoscopic Image Desmoking
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.09849