Command Palette
Search for a command to run...
Huang Zhida ; Zhong Zhuoyao ; Sun Lei ; Huo Qiang
Abstract
In this paper, we present a new Mask R-CNN based text detection approachwhich can robustly detect multi-oriented and curved text from natural sceneimages in a unified manner. To enhance the feature representation ability ofMask R-CNN for text detection tasks, we propose to use the Pyramid AttentionNetwork (PAN) as a new backbone network of Mask R-CNN. Experiments demonstratethat PAN can suppress false alarms caused by text-like backgrounds moreeffectively. Our proposed approach has achieved superior performance on bothmulti-oriented (ICDAR-2015, ICDAR-2017 MLT) and curved (SCUT-CTW1500) textdetection benchmark tasks by only using single-scale and single-model testing.
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| scene-text-detection-on-icdar-2015 | PAN | F-Measure: 85.9 Precision: 90.8 Recall: 81.5 |
| scene-text-detection-on-icdar-2017-mlt-1 | PAN | F-Measure: 74.3% Precision: 80 Recall: 69.8 |
| scene-text-detection-on-scut-ctw1500 | PAN | F-Measure: 85 FPS: 65.2 Precision: 86.8 Recall: 83.2 |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.