Scene Text Detection with Supervised Pyramid Context Network

21 Nov 2018  ·  Enze Xie, Yuhang Zang, Shuai Shao, Gang Yu, Cong Yao, Guangyao Li ·

Scene text detection methods based on deep learning have achieved remarkable results over the past years. However, due to the high diversity and complexity of natural scenes, previous state-of-the-art text detection methods may still produce a considerable amount of false positives, when applied to images captured in real-world environments. To tackle this issue, mainly inspired by Mask R-CNN, we propose in this paper an effective model for scene text detection, which is based on Feature Pyramid Network (FPN) and instance segmentation. We propose a supervised pyramid context network (SPCNET) to precisely locate text regions while suppressing false positives. Benefited from the guidance of semantic information and sharing FPN, SPCNET obtains significantly enhanced performance while introducing marginal extra computation. Experiments on standard datasets demonstrate that our SPCNET clearly outperforms start-of-the-art methods. Specifically, it achieves an F-measure of 92.1% on ICDAR2013, 87.2% on ICDAR2015, 74.1% on ICDAR2017 MLT and 82.9% on Total-Text.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Scene Text Detection ICDAR 2013 SPCNET F-Measure 92.1% # 2
Precision 93.8 # 4
Recall 90.5 # 3
Scene Text Detection ICDAR 2015 SPCNET F-Measure 87.2 # 15
Precision 88.7 # 22
Recall 85.8 # 13
Scene Text Detection ICDAR 2017 MLT SPCNET Precision 80.6 # 7
Recall 68.6 # 9
F-Measure 74.1% # 6
Scene Text Detection Total-Text SPCNET F-Measure 82.9% # 17
Precision 83 # 17
Recall 82.8 # 8