AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

PMSPO: Progressive Matching and Semantic-Aware Policy Optimization for Camouflaged Object Detection

Maosheng Su, Junlei Cao, Zhiyuan Wang, Shuo Wang, Ning Zhang, Jun Luo

ICML 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection

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 Sinkhorn matching instead of greedy matching (ablation)PMSPO reward · COD10KF-measure (Fβ): 0.827 -> 0.839Improved
use Sinkhorn matching instead of Hungarian matching (ablation)PMSPO reward · COD10KF-measure (Fβ): 0.830 -> 0.839Improved
use Sinkhorn matching instead of greedy matching (ablation)PMSPO reward · CHAMELEONMAE (M): 0.054 -> 0.048Improved
add PLGF filtering with delta 0.03 (ablation)Sinkhorn reward · COD10KMAE (M): 0.035 -> 0.029Improved
add PLGF filtering with delta 0.03 (ablation)Sinkhorn reward · CHAMELEONMAE (M): 0.048 -> 0.044Improved
raise PLGF threshold from 0.03 to 0.05 (ablation)PLGF · COD10KMAE (M): 0.029 -> 0.031Hurt
raise PLGF threshold from 0.03 to 0.1 (ablation)PLGF · COD10KMAE (M): 0.029 -> 0.032Hurt
add semantic contrastive reward without curriculum (ablation)Sinkhorn + PLGF · COD10KMAE (M): 0.029 -> 0.026Mixed
add semantic contrastive reward without curriculum (ablation)Sinkhorn + PLGF · COD10KSα: 0.863 -> 0.867No clear effect
add curriculum reward framework (ablation)Sinkhorn + PLGF + contrastive reward · COD10KMAE (M): 0.026 -> 0.024Improved
add curriculum reward framework (ablation)Sinkhorn + PLGF + contrastive reward · CAMOMAE (M): 0.066 -> 0.057Improved
add curriculum reward framework (ablation)Sinkhorn + PLGF + contrastive reward · CHAMELEONFβ: 0.862 -> 0.875Improved

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