AI Paper Atlas

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

Locate Then Segment: A Strong Pipeline for Referring Image Segmentation

Ya Jing, Tao Kong, Wei Wang, Liang Wang, Lei Li, Tieniu Tan

CVPR 2021 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 154 citations (OpenAlex)

Check 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
Decouple referring segmentation into localization followed by mask refinementLTS · RefCOCO+ and RefCOCOgIoU: +3.2% on RefCOCO+ and +3.4% on RefCOCOgImproved
Add cross-modal fusion (ablation)LTS without fusion · RefCOCO valIoU: 56.94 -> 65.05Improved
Add language-guided relevance filtering (ablation)LTS without filter · RefCOCO valIoU: 63.50 -> 65.05Improved
Add the segmentation refinement module (ablation)LTS without segmentation · RefCOCO valIoU: 56.94 -> 65.05Improved
Add locating loss (ablation)LTS without locating loss · RefCOCO valIoU: 65.05 -> 65.43Improved
Add LTS segmentation module to CMPC (ablation)CMPC · RefCOCO val, testA, testBIoU: 61.36 -> 62.75; 64.53 -> 65.34; 59.64 -> 61.08Improved
Replace relevance filtering with transformer localization (ablation)LTS · RefCOCO, RefCOCO+, RefCOCOg valIoU: 65.43 -> 66.15; 54.21 -> 54.52; 54.40 -> 54.51Improved
Apply relevance filtering twice instead of once (ablation)LTS · RefCOCO val, testA, testBIoU: 65.43 -> 66.04; 67.76 -> 68.68; 63.08 -> 63.27Improved
Apply relevance filtering three times instead of once (ablation)LTS · RefCOCO valIoU: 65.43 -> 65.54No clear effect
Generate the filtering kernel from each word rather than the sentence feature (ablation)LTS · RefCOCO val, testA, testBIoU: 65.43 -> 64.92; 67.76 -> 67.31; 63.08 -> 62.50Hurt
Increase input resolution (ablation)LTS · RefCOCO val, testA, testBIoU: 63.01 -> 65.90; 65.40 -> 68.16; 60.78 -> 63.45Improved
Use transformer, more filtering times, and larger input resolution (ablation)LTS · RefCOCO, RefCOCO+, RefCOCOg valIoU: 65.43 -> 66.75; 54.21 -> 54.94; 54.40 -> 54.51Improved

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