AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Synthetically Supervised Feature Learning for Scene Text Recognition

Yang Liu, Zhaowen Wang, Hailin Jin, Ian Wassell

ECCV 2018 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 87 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
add ideal rendered clean image as auxiliary supervision (ablation)CRNN · SVTaccuracy (%): 80.8 -> 87.0Improved
use binarized image as clean target instead of none (ablation)CRNN · SVTaccuracy (%): 80.8 -> 85.8Improved
use deskewed image as clean target instead of none (ablation)CRNN · SVTaccuracy (%): 80.8 -> 84.7Improved
use ideal image instead of binarized image as clean target (ablation)synthetically supervised model · SVTaccuracy (%): 85.8 -> 87.0Improved
add image generation loss Lg (ablation)CRNN · SVTaccuracy (%): 80.8 -> 86.1Improved
add feature consistency loss Lf (ablation)CRNN · SVTaccuracy (%): 80.8 -> 85.1Improved
add image discriminator adversarial loss Lga to image generation (ablation)Image Generation variant · SVTaccuracy (%): 86.1 -> 84.7Hurt
add feature discriminator adversarial loss Lfa plus Lg and Lf (ablation)Feature Matching variant · SVTaccuracy (%): 85.1 -> 87.0Improved
add synthetic clean-image supervision (full model vs CRNN) on irregular textCRNN · SVT-Perspectiveaccuracy (%): 66.8 -> 73.9Improved
add synthetic clean-image supervision (full model vs CRNN) on curved textCRNN · CUTE80accuracy (%): 54.9 -> 62.5Improved
add synthetic clean-image supervision on lexicon-free IIIT5KCRNN · IIIT5K (lexicon-free)accuracy (%): 78.2 -> 89.4Improved

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