AI Paper Atlas

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

Direct Segmentation without Logits Optimization for Training-Free Open-Vocabulary Semantic Segmentation

Jiahao Li, Yang Lu, Yachao Zhang, Fangyong Wang, Yuan Xie, Yanyun Qu

CVPR 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 0 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 KL divergence map between raw logits and degenerate distribution (ablation)raw CLIP logits baseline · 4-benchmark averagemIoU: 8.9 -> 17.8Improved
add non-maximum suppression on raw logits (ablation)KL discrepancy pipeline · 4-benchmark averagemIoU: 17.8 -> 19.1Improved
replace bilinear upsampling with joint bilateral upsampling (ablation)KL discrepancy pipeline · 4-benchmark averagemIoU: 19.1 -> 19.6Improved
replace KL map with optimal path map (ablation)KL + NMS + JBU pipeline · 4-benchmark averagemIoU: 19.6 -> 40.0Improved
replace KL map with maximum velocity map (ablation)KL + NMS + JBU pipeline · 4-benchmark averagemIoU: 19.6 -> 40.8Improved
fuse optimal path and maximum velocity maps (ablation)maximum velocity mode · 4-benchmark averagemIoU: 40.8 -> 38.4Hurt
use maximum velocity mode instead of optimal path modeOurs · OVSS 8-benchmark average (CLIP ViT-B/16)mIoU: 46.2 -> 46.9Improved
combine SD2 up0 and up1 attention blocks instead of up1 alone (ablation)maximum velocity mode · 4-benchmark averagemIoU: 39.7 -> 40.8Improved
use key-key mean attention for logits instead of original attention (ablation)CLIP logits · 4-benchmark averagemIoU: 36.2 -> 40.8Improved
use key-key instead of query-query attention for logits (ablation)CLIP logits · 4-benchmark averagemIoU: 39.1 -> 40.8Improved
raise threshold tau above 0.3 (ablation)maximum velocity mode · five benchmarksmIoU: peak at tau=0.3Hurt
use more SD2 denoising steps (ablation)maximum velocity mode · five benchmarksmIoUHurt

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