AI Paper Atlas

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

Towards Unified Scene Text Spotting Based on Sequence Generation

Taeho Kil, Seonghyeon Kim, Sukmin Seo, Yoonsik Kim, Daehee Kim

CVPR 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 31 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 starting-point prompting at test time (iterative re-prompting from last detected text) (ablation)UNITS on TextOCR validation · TextOCR validationEnd-to-end F-measure: 44.0 -> 64.7Improved
add starting-point prompting (ablation)UNITS on TextOCR validation · TextOCR validationEnd-to-end Recall: 30.6 -> 54.2Improved
reduce decoder length from 1024 to 256 with starting-point prompting (ablation)UNITS on TextOCR validation · TextOCR validationEnd-to-end F-measure: 64.7 -> 61.8Mixed
replace Swin encoder with hybrid ResNet-50 + transformer encoder (ablation)UNITS · ICDAR 2015 and Total-TextEnd-to-end E2E: 78.5 -> 65.6 (ICDAR 2015); 77.3 -> 67.4 (Total-Text)Hurt
add multi-way (detection/recognition expert FFN) decoder (ablation)UNITS decoder · ICDAR 2015End-to-end E2E: 39.1 -> 40.7Improved
add multi-way (detection/recognition expert FFN) decoder (ablation)UNITS decoder · Total-TextEnd-to-end E2E: 45.1 -> 47.7Improved
add multi-way decoder (ablation)UNITS decoder · ICDAR 2015Detection: 73.7 -> 75.1Improved
add multi-way decoder (ablation)UNITS decoder · Total-TextDetection: 76.1 -> 78.9Improved
unified multi-format training with a single shared model instead of per-dataset fine-tuned modelsUNITS · ICDAR 2015 End-to-end StrongEnd-to-end Strong F-measure: 89.0 -> 88.4Hurt
unified multi-format model vs dataset-specific fine-tuning on Total-TextUNITS · Total-TextEnd-to-end Full F-measure: 86.0 -> 85.0Hurt
use the sequence generation model with starting-point prompting and unified multi-format interface, compared to SPTS on ICDAR 2015 point metricUNITS_Shared (quad) vs SPTS · ICDAR 2015 (point-based metric)End-to-end Strong: 77.5 -> 89.9Improved

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