AI Paper Atlas

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

Knowledge Mining With Scene Text for Fine-Grained Recognition

Hao Wang, Junchao Liao, Tianheng Cheng, Zewen Gao, Hao Liu, Bo Ren, Xiang Bai, Wenyu Liu

CVPR 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 19 citations (OpenAlex)

Check 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 KnowBert knowledge-enhanced text features to ViTViT image classifier · Con-TextmAP: 79.24 -> 89.53Improved
Add KnowBert knowledge-enhanced text features to ViTViT image classifier · Drink BottlemAP: 80.81 -> 85.26Improved
Add KnowBert knowledge-enhanced text features to ViTViT image classifier · Crowd ActivitymAP: 80.29 -> 87.45Improved
Add KARC to the ViT + BERT baseline (ablation)ViT + BERT · Con-TextmAP: 86.51 -> 87.25Improved
Add KARC to the ViT + BERT baseline (ablation)ViT + BERT · Drink BottlemAP: 82.81 -> 83.59Improved
Add KARC to the ViT + BERT baseline (ablation)ViT + BERT · Crowd ActivitymAP: 85.34 -> 86.16Improved
Add VKAC on top of KARC (ablation)Model A with KARC · Con-TextmAP: 87.25 -> 89.53Improved
Add VKAC on top of KARC (ablation)Model A with KARC · Drink BottlemAP: 83.59 -> 85.26Improved
Add VKAC on top of KARC (ablation)Model A with KARC · Crowd ActivitymAP: 86.16 -> 87.45Improved
Train the full model end-to-end instead of separately (ablation)ViT + KnowBert classifier · Con-TextmAP: 81.47 -> 89.53Improved
Train the full model end-to-end instead of separately (ablation)ViT + KnowBert classifier · Drink BottlemAP: 82.26 -> 85.26Improved
Train the full model end-to-end instead of separately (ablation)ViT + KnowBert classifier · Crowd ActivitymAP: 81.79 -> 87.45Improved

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