atlas

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

BTS: A Bi-Lingual Benchmark for Text Segmentation in the Wild

CVPR 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 16 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Train with human-labeled BTS alongside TextSeg rather than TextSeg alonePGTSNet(base) · BTS testfgIoU: 64.13 -> 83.68Helped
Use TextSeg and BTS instead of TextSeg and synthetic bilingual dataPGTSNet(base) · BTS testfgIoU: 74.59 -> 83.68Helped
Add the text perceptual discriminator (ablation)PGTSNet(base) · BTS validationfgIoU: 83.68 -> 84.78Helped
Add the TV loss to the discriminator variant (ablation)PGTSNet · BTS validationfgIoU: 84.78 -> 84.93Helped
Add the detection module to the discriminator and TV-loss variant (ablation)PGTSNet · BTS validationfgIoU: 84.93 -> 86.48Helped
Use PGTSNet(final) instead of the strongest listed baseline for fgIoUBTS text segmentation · BTS testfgIoU: 82.76 -> 86.48Helped
Use PGTSNet(final) instead of the strongest listed baseline for F-scoreBTS text segmentation · BTS testF-score: 89.24 -> 90.98Helped

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