AI Paper Atlas

Research map / Detection and segmentation / CNN features

MANGO: A Mask Attention Guided One-Stage Scene Text Spotter

AAAI 2021 · Detection and segmentation › CNN features · approach: Arbitrary-shape scene text detection · 80 citations (OpenAlex)

Read the paperCheck 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
Increase grid parameter S from 20 to 40 (ablation)MANGO · IC13End-to-End S: 83.2 -> 90.5Helped
Increase grid parameter S from 20 to 60 (ablation)MANGO · IC15End-to-End S: 33.8 -> 81.8Helped
Increase grid parameter S from 20 to 50 (ablation)MANGO · Total-TextEnd-to-End None: 46.9 -> 73.1Helped
Replace original position annotations with rectangular bounding boxes (ablation)MANGO · IC15End-to-End G: 67.3 -> 64.0Hurt
Replace original position annotations with rectangular bounding boxes (ablation)MANGO · Total-TextEnd-to-End None: 72.9 -> 69.7Hurt
Use larger input scale (longer side 1600 instead of 1280)MANGO · Total-TextEnd-to-End None: 71.7 -> 72.9Helped
Use larger input scale (longer side 1600 instead of 1280)MANGO · Total-TextFPS: 8.9 -> 4.3Hurt
Use character-weighted grid voting to combine recognition outputsMANGO · Scene text spottingrecognition reliabilityHelped

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