AI Paper Atlas

Research map / Detection and segmentation / CNN features

Learning Spatial-Semantic Context with Fully Convolutional Recurrent Network for Online Handwritten Chinese Text Recognition

Zecheng Xie, Zenghui Sun, Lianwen Jin, Hao Ni, Terry Lyons

TPAMI 2017 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 132 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
raise path signature truncation from Sig0 to Sig3 (ablation)FCRN · Dataset-ICDARcorrect rate (CR, %): 90.18 -> 92.35Improved
raise path signature truncation from Sig0 to Sig3 (ablation)FCRN · Dataset-CASIAcorrect rate (CR, %): 92.59 -> 94.52Improved
raise path signature from Sig2 to Sig3 (ablation)FCRN · Dataset-ICDARcorrect rate (CR, %): 92.25 -> 92.35Improved
add a second receptive-field scale (2C-FCRN) (ablation)FCRN · Dataset-ICDARcorrect rate (CR, %): 92.25 -> 93.08Improved
add a third receptive-field scale (3C-FCRN) (ablation)FCRN · Dataset-ICDARcorrect rate (CR, %): 92.25 -> 93.53Improved
add three receptive-field scales (3C-FCRN) (ablation)FCRN · Dataset-CASIAcorrect rate (CR, %): 94.37 -> 95.16Improved
use same-scale receptive fields with similar size (FCRN-2) (ablation)FCRN · Dataset-ICDARcorrect rate (CR, %): 92.25 -> 92.40No clear effect
use same-scale receptive fields with similar size (FCRN-3) (ablation)FCRN · Dataset-CASIAcorrect rate (CR, %): 94.37 -> 94.17No clear effect
add bidirectional rather than unidirectional LSTM in implicit LM (SLD) (ablation)3C-FCRN · Dataset-ICDARcorrect rate (CR, %): 94.59 -> 95.19Improved
add implicit language model (B_SLD)3C-FCRN · Dataset-ICDARcorrect rate (CR, %): 93.53 -> 95.19Improved
add trigram LM on top of implicit LM (ablation)3C-FCRN+B_SLD · Dataset-ICDARcorrect rate (CR, %): 95.19 -> 97.15Improved
use larger SLD corpus rather than PFR for implicit LM (ablation)3C-FCRN+B · Dataset-ICDARcorrect rate (CR, %): 94.90 -> 95.19Improved

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