AI Paper Atlas

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

CorrCLIP: Reconstructing Patch Correlations in CLIP for Open-Vocabulary Semantic Segmentation

Dengke Zhang, Fagui Liu, Quan Tang

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 26 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 scope reconstruction (SAM masks) (ablation)ClearCLIP baseline · VOC21mIoU: 53.6 -> 68.1Improved
add value reconstruction (DINO similarity) (ablation)ClearCLIP + SR · VOC21mIoU: 68.1 -> 68.5Improved
add map correction (ablation)ClearCLIP + SR + VR · VOC21mIoU: 68.5 -> 72.5Improved
add feature refinement (ablation)ClearCLIP + SR + VR + MC · VOC21mIoU: 72.5 -> 74.8Improved
add scope reconstruction (ablation)SCLIP · VOC21mIoU: 59.6 -> 66.5Improved
add scope reconstruction (ablation)ProxyCLIP · VOC21mIoU: 61.3 -> 67.0Improved
add scope reconstruction (ablation)SC-CLIP · VOC21mIoU: 64.6 -> 66.0Improved
increase SR final mask size from 21x21 to 84x84 (ablation)CorrCLIP SR · VOC21mIoU: 64.7 -> 70.2Improved
use DINO QK-QK similarity instead of CLIP Q-K (ablation)CorrCLIP value reconstruction · VOC21mIoU: 69.5 -> 74.2Improved
use DINO QK-QK similarity instead of uniform similarity (ablation)CorrCLIP value reconstruction · VOC21mIoU: 73.4 -> 74.2Improved
add mask merging (clustering) (ablation)CorrCLIP · VOC21mIoU: 74.2 -> 74.8Improved
add semantic branch alone (no spatial branch) (ablation)CorrCLIP · CityscapesmIoU: 48.3 -> 47.6Hurt

Later papers that compared with this method

4 later accepted papers report CorrCLIP as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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