AI Paper Atlas

Research map / Detection and segmentation / CNN features

PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text

Wenhai Wang, Enze Xie, Xiang Li, Xuebo Liu, Ding Liang, Zhibo Yang, Tong Lü, Chunhua Shen

TPAMI 2021 · Detection and segmentation › CNN features · approach: Arbitrary-shape scene text detection · 162 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
stack 1 FPEM instead of none (ablation)PAN++ · Total-Text spottingend-to-end F-measure: 64.5 -> 66.0Improved
stack 2 FPEMs instead of none (ablation)PAN++ · Total-Text spottingend-to-end F-measure: 64.5 -> 66.4Improved
stack 4 FPEMs instead of 2 (ablation)PAN++ · Total-Text spottingend-to-end F-measure: 66.4 -> 66.6No clear effect
stack 2 FPEMs instead of none (ablation)PAN++ · IC15 spottingend-to-end F-measure: 63.7 -> 64.7Improved
stack 4 FPEMs instead of 2 (ablation)PAN++ · Total-Text spottingFPEM time (ms): 3.8 -> 6.5Hurt
add improved Pixel Aggregation (background term) to PAN (ablation)PAN · Total-Text detectionF-measure: 83.5 -> 84.0Improved
add improved FPEM to PAN + PA (impr) (ablation)PAN · Total-Text detectionF-measure: 84.0 -> 84.5Improved
add improved Pixel Aggregation to PAN (ablation)PAN · IC15 detectionF-measure: 80.3 -> 80.7Improved
add improved FPEM to PAN + PA (impr) (ablation)PAN · IC15 detectionF-measure: 80.7 -> 81.2Improved
fine-tune on SynthText before Total-TextPAN++ detection · Total-Text detectionF-measure: 84.5 -> 85.3Improved
joint training on multiple datasets instead of fine-tuningPAN++ (S:736) · Total-Text spottingend-to-end F-measure: 67.1 -> 68.6Improved
raise shrinking rate to 1 (kernel equals text region) (ablation)PAN++ detection · Total-Text and IC15F-measureHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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