atlas

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

Char-Net: A Character-Aware Neural Network for Distorted Scene Text Recognition

AAAI 2018 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 183 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
add hyper-connections to word-level encoder (ablation)Char-Net[N] · IC-ISTaccuracy: 72.6 -> 74.2Helped
add hyper-connections to word-level encoder (ablation)Char-Net[N] · SVT-Paccuracy: 75.7 -> 78.9Helped
replace traditional attention with hierarchical attention (no CLE) (ablation)attention recognizer · IC-ISTaccuracy: 70.1 -> 70.8Helped
add character-level encoder with local STN on top of HAM (ablation)Char-Net · IC-ISTaccuracy: 70.8 -> 74.2Helped
add full Char-Net components (ablation)Char-Net · IC-03/IC-13/SVT/IIIT5KaccuracyNo clear effect
character-level local rectification vs global TPS rectificationscene text recognizer · SVTaccuracy: 83.6 -> 84.4Helped

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