AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Frozen CLIP-DINO: A Strong Backbone for Weakly Supervised Semantic Segmentation

Bingfeng Zhang, Siyue Yu, Jimin Xiao, Yunchao Wei, Yao Zhao

TPAMI 2025 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised semantic segmentation · 34 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 RFM CAM refinement module to the decoder (ablation)WeCLIP · PASCAL VOC 2012 val (w/o CRF)mIoU: 68.7 -> 74.9Improved
use 3 transformer layers in decoder instead of 1 (ablation)WeCLIP decoder · PASCAL VOC 2012 val (w/o CRF)mIoU: 73.2 -> 74.9Improved
use 5 decoder transformer layers instead of 3 (ablation)WeCLIP decoder · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.9 -> 70.3Hurt
add decoder affinity A_f to static attention refinement A_s (ablation)RFM refining map · PASCAL VOC 2012 val (w/o CRF)mIoU: 71.8 -> 74.3Improved
add reliability filter G_e to A_s (ablation)RFM refining map · PASCAL VOC 2012 val (w/o CRF)mIoU: 71.8 -> 72.3Improved
add filter G_e to A_f + A_s (ablation)RFM refining map · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.3 -> 74.9Improved
use frozen CLIP features from all 12 blocks instead of only block 12 (ablation)WeCLIP decoder input · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.3 -> 74.9Improved
supervise A_f with online pseudo label M_p instead of argmax of prediction P (ablation)WeCLIP affinity loss · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.6 -> 74.9Improved
supervise A_f with intersection of M_p and P instead of M_p (ablation)WeCLIP affinity loss · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.9 -> 74.8No clear effect
add affinity loss weight lambda=0.1 versus 0 (ablation)WeCLIP loss · PASCAL VOC 2012 val (w/o CRF)mIoU: 73.3 -> 74.9Improved
increase lambda from 0.1 to 1.0 (ablation)WeCLIP loss · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.9 -> 72.9Hurt
use multi-scale {0.75, 1.0} instead of single scale {1.0} at inference (ablation)WeCLIP inference · PASCAL VOC 2012 val (w/o CRF)mIoU: 74.0 -> 74.9Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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