AI Paper Atlas

Research map / Detection and segmentation / CNN features

RHCNet: Residual-Guided Hierarchical Calibration Network for Robust Underwater Object Detection

Yueying Wang, Yiteng Guo, Weidong Zhang, Jie Wen, Liquan Shen, Huaicheng Yan, Xin Xu

CVPR 2026 · Detection and segmentation › CNN features · approach: Efficient DETR query design

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 LAM (ablation)RHCNet (ResNet-50) · DUOAP: 66.45 -> 70.53Improved
add RGFE (ablation)RHCNet (ResNet-50) · DUOAP: 66.90 -> 70.53Improved
add PAM (ablation)RHCNet (ResNet-50) · DUOAP: 66.74 -> 70.53Improved
add CGCA (ablation)RHCNet (ResNet-50) · DUOAP: 65.86 -> 70.53Improved
add PAM and CGCA together (ablation)RHCNet (ResNet-50) · DUOAP: 65.24 -> 70.53Improved
add LAM and RGFE together (ablation)RHCNet (ResNet-50) · DUOAP: 65.26 -> 70.53Improved
use stronger ResNet-101 backbone with FPN and all modules (ablation)RHCNet · DUOAP: 70.53 -> 70.03No clear effect
replace BiFPN neck with HFCP (ablation)RHCNet (ResNet-50) · DUOAP: 66.45 -> 70.53Improved
replace vanilla self-attention with cluster-guided CGCA (ablation)RHCNet (ResNet-50) · DUOAP: 68.12 -> 70.53Improved
full RHCNet over RetinaNet baseline (ablation)RetinaNet R-50 FPN · DUOAP: 57.06 -> 70.53Improved
apply RHCNet to non-underwater dataRHCNet · COCOAP: 45.68 vs 44.18 (YOLOv11)Improved

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