AI Paper Atlas

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

ABINet++: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Spotting

Shancheng Fang, Zhendong Mao, Hongtao Xie, Yuxin Wang, Chenggang Clarence Yan, Yongdong Zhang

TPAMI 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 95 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 position attention and 3 Transformer layers (LV) over small vision model (ablation)vision model · STR averageavg accuracy: 88.8 -> 89.8Improved
pre-train vision model (PVM) (ablation)LV+BCN · STR averageavg accuracy: 91.7 -> 92.3Improved
add LM pre-training on WikiText-103 (PLM_out) (ablation)LV+BCN · STR averageavg accuracy: 92.3 -> 92.5Improved
allow gradient flow between vision and language models (ablation)LV+BCN · STR averageavg accuracy: 92.3 -> 91.4Hurt
bidirectional BCN instead of unidirectional SRN-U (ablation)SV · STR averageavg accuracy: 90.2 -> 91.0Improved
BCN instead of SRN (ablation)LV · STR averageavg accuracy: 91.9 -> 92.3Improved
BCN instead of SRN in spelling correction (ablation)language model · text spelling correctionword accuracy: 27.6 -> 41.9Improved
BCN instead of BERT (ablation)LV · STR averageavg accuracy: 92.3 -> 92.3No clear effect
iterate BCN 3 times (ablation)SV+BCN · STR averageavg accuracy: 91.0 -> 91.4Improved
iterate BCN 3 times (ablation)LV+BCN · STR averageavg accuracy: 92.3 -> 92.6Improved
add language model fine-tuned on target data (ablation)spotting · Total-TextHmean: 70.2 -> 76.3Improved
add language model fine-tuned on target data (ablation)spotting · SCUT-CTW1500Hmean: 53.5 -> 59.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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