Papers › Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation

Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation

25 Feb 2018CVPR 2018 6arXiv:1802.08948archive 2025-07-28

Pengyuan Lyu, Cong Yao, Wenhao Wu, Shuicheng Yan, Xiang Bai

Previous deep learning based state-of-the-art scene text detection methods can be roughly classified into two categories. The first category treats scene text as a type of general objects and follows general object detection paradigm to localize scene text by regressing the text box locations, but troubled by the arbitrary-orientation and large aspect ratios of scene text. The second one segments text regions directly, but mostly needs complex post processing. In this paper, we present a method that combines the ideas of the two types of methods while avoiding their shortcomings. We propose to detect scene text by localizing corner points of text bounding boxes and segmenting text regions in relative positions. In inference stage, candidate boxes are generated by sampling and grouping corner points, which are further scored by segmentation maps and suppressed by NMS. Compared with previous methods, our method can handle long oriented text naturally and doesn't need complex post processing. The experiments on ICDAR2013, ICDAR2015, MSRA-TD500, MLT and COCO-Text demonstrate that the proposed algorithm achieves better or comparable results in both accuracy and efficiency. Based on VGG16, it achieves an F-measure of 84.3% on ICDAR2015 and 81.5% on MSRA-TD500.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

lvpengyuan/corner mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Multi-Oriented Scene Text DetectionObject DetectionScene Text DetectionText Detectionobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Scene Text Detection ICDAR 2013 Corner Localization (multi-scale) F-Measure 88% #7 of 16 Archive leaderboard report
Scene Text Detection ICDAR 2013 Corner Localization (multi-scale) Precision 92 #7 of 16 Archive leaderboard report
Scene Text Detection ICDAR 2013 Corner Localization (multi-scale) Recall 84.4 #7 of 16 Archive leaderboard report
Scene Text Detection ICDAR 2015 Corner Localization (multi-scale) F-Measure 84.3 #27 of 43 Archive leaderboard report
Scene Text Detection ICDAR 2015 Corner Localization (multi-scale) Precision 89.5 #27 of 43 Archive leaderboard report
Scene Text Detection ICDAR 2015 Corner Localization (multi-scale) Recall 79.7 #27 of 43 Archive leaderboard report
Scene Text Detection ICDAR 2017 MLT Corner Localization (single-scale) F-Measure 66.8% #2 of 14 Archive leaderboard report
Scene Text Detection ICDAR 2017 MLT Corner Localization (single-scale) Precision 83.8 #2 of 14 Archive leaderboard report
Scene Text Detection ICDAR 2017 MLT Corner Localization (single-scale) Recall 55.6 #2 of 14 Archive leaderboard report
Scene Text Detection ICDAR 2017 MLT Corner Localization (multi-scale) F-Measure 72.4% #14 of 14 Archive leaderboard report
Scene Text Detection ICDAR 2017 MLT Corner Localization (multi-scale) Precision 74.3 #14 of 14 Archive leaderboard report
Scene Text Detection ICDAR 2017 MLT Corner Localization (multi-scale) Recall 70.6 #14 of 14 Archive leaderboard report
Scene Text Detection MSRA-TD500 Corner Localization F-Measure 81.5 #13 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 Corner Localization Precision 87.6 #13 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 Corner Localization Recall 76.2 #13 of 18 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections