AI Paper Atlas

Research map / Detection and segmentation / CNN features

HUPE: Heuristic Underwater Perceptual Enhancement with Semantic Collaborative Learning

Zengxi Zhang, Zhiying Jiang, Long Ma, Jinyuan Liu, Xin Fan, Risheng Liu

IJCV 2025 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 36 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
add style loss (ablation)HUPE · UIEBDvisual contrast / quantitative metricsImproved
add L1 and contrastive losses (ablation)HUPE · UIEBDvisual color fidelityImproved
add gradient and depth maps as inputs to the heuristic prior guided encoder (ablation)HUPE · underwater scenesbrightness and contrastImproved
include RGB image as encoder input (ablation)HUPE · underwater scenesvisual brightnessImproved
add phase branch in spatial-frequency affine block (ablation)HUPE · underwater scenesscene reflection recoveryImproved
add amplitude branch in spatial-frequency affine block (ablation)HUPE · underwater scenesoverall brightnessImproved
add frequency components (both amplitude and phase) (ablation)HUPE · underwater scenescolor castImproved
use 3 hybrid invertible blocks instead of fewer than 3 (ablation)HUPE · underwater scenesall evaluation metrics: N<3 -> N=3Improved
increase hybrid invertible blocks beyond 3 (ablation)HUPE · underwater scenesimage qualityNo clear effect
use 6 flow steps instead of fewer than 4 (ablation)HUPE · underwater scenescolor balanceImproved
increase flow steps beyond 6 (ablation)HUPE · underwater scenesquantitative metricsNo clear effect
add semantic collaborative learning instead of traditional cascade (HUPE-CS) (ablation)HUPE · RUIE/Aquarium detection and SUIM segmentationAP / IoUImproved

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