AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognition

Zhi Qiao, Yu Zhou, Dongbao Yang, Yucan Zhou, Weiping Wang

CVPR 2020 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 320 citations (OpenAlex)

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 word embedding supervision alone (ablation)ASTER · IC13, SVTP, and IC15accuracy: IC13 90.9 -> 90.8; SVTP 79.1 -> 79.2; IC15 78.4 -> 77.0No clear effect
Initialize the decoder with predicted holistic features without word embedding supervision (ablation)ASTER · IC13, SVTP, and IC15accuracy: IC13 90.9 -> 91.1; SVTP 79.1 -> 78.1; IC15 78.4 -> 76.1Mixed
Combine word embedding supervision with predicted semantic decoder initialization (ablation)ASTER · IC13, SVTP, and IC15accuracy: +1.9%; +2.3%; +1.6%Improved
Add word embedding supervision to ASTER with semantic initialization (ablation)ASTER + semantic initialization · IC13, SVTP, and IC15accuracy: +1.7%; +3.3%; +3.9%Improved
Use FastText word embedding to initialize the decoder instead of predicted semantics (ablation)SE-ASTER · Scene text recognitionaccuracyNo clear effect
Predict semantic information from the final BiLSTM hidden state rather than the BiLSTM outputs (ablation)SE-ASTER · Scene text recognitionaccuracyHurt
Add semantic prediction and decoder initializationSAR · IC13accuracy: 91.0 -> 90.9Hurt
Add semantic prediction and decoder initializationSAR · IC15, SVT, and SVTPaccuracy: +4.2%; +1.3%; +2.3%Improved
Use SE-ASTER rather than ASTERASTER · IC15 and SVTPaccuracy: IC15 76.1% -> 80.0%; SVTP 78.5% -> 81.4%Improved
Use SE-ASTER rather than ASTERASTER · IIIT5K and CUTEaccuracy: IIIT5K +0.4%; CUTE +4.1%Improved
Use SE-ASTER rather than ASTERASTER · IC13 and SVTaccuracy: IC13 92.8%; SVT 89.6%Mixed
Use semantic prediction with word-embedding supervision and decoder guidanceASTER · IC13-sr and IC15-sraccuracy: IC13-sr 71.4 -> 77.4; IC15-sr 65.6 -> 70.0Improved

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