AI Paper Atlas

Research map / Detection and segmentation / CNN features

SEE: Towards Semi-Supervised End-to-End Scene Text Recognition

Christian Bartz, Haojin Yang, Christoph Meinel

AAAI 2018 · Detection and segmentation › CNN features · approach: Arbitrary-shape scene text detection · 67 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
Use a ResNet localization network instead of VGGNet (ablation)SEE · Scene text localization experimentstraining speed and successImproved
Apply rotation dropout to affine transformation parameters (ablation)SEE localization network · Scene text localizationexcessive rotation in predicted transformationsImproved
Add localization-specific regularizers for grid area, aspect ratio, and mirrored grids (ablation)SEE · Scene text localizationconvergence speedImproved
Train with a curriculum that gradually increases image complexity (ablation)SEE · SVHN images with multiple scattered text linestraining convergenceImproved
Restart training with the localization network's previous weights and a randomly initialized recognition network (ablation)SEE · Scene text localization and recognitionoverall performanceImproved
Use ResNet-18 for recognition on FSNS (ablation)SEE · FSNSrecognition accuracyImproved

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