AI Paper Atlas

Research map / Image generation / Image restoration

Benchmarking Endoscopic Surgical Image Restoration and Beyond

Jialun Pei, Diandian Guo, Donghui Yang, Zhixi Li, Yuxin Feng, Long Ma, Bo Du, Pheng-Ann Heng

CVPR 2026 · Image generation › Image restoration · approach: Reflection removal · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use task-specific desmoking model MS-CNN instead of generic ConvIRSurgClean desmoking · SurgClean DesmokingPSNR (dB): 19.43 -> 19.02Hurt
use task-specific defogging models instead of generic restoration modelsSurgClean defogging · SurgClean DefoggingPSNR (dB)Hurt
use ConvIR instead of Histoformer for restoration before depth estimationDepth estimation on defogged images · SurgClean DefoggingAbs Rel: 0.4136 -> 0.3302Improved
use MambaIRv2 instead of AST for restoration before semantic parsingSemantic parsing on defogged images · SurgClean DefoggingmIoU: 0.2600 -> 0.4204Improved
use Histoformer instead of ConvIR on desmokingSurgClean desmoking · SurgClean DesmokingPSNR (dB): 19.432 -> 17.666Hurt
use AST instead of Restormer on desplashingSurgClean desplashing · SurgClean DesplashingPSNR (dB): 21.396 -> 22.046Improved
use optical-flow warping of unaligned references during training and evaluationAll benchmarked models · SurgCleanPSNR (dB)No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy