ICDAR2017 Robust Reading Challenge on COCO-Text
Research output: Contribution to journal › Conference article › Research › peer-review
This report presents the final results of the ICDAR 2017 Robust Reading Challenge on COCO-Text. A challenge on scene text detection and recognition based on the largest real scene text dataset currently available: the COCO-Text dataset. The competition is structured around three tasks: Text Localization, Cropped Word Recognition and End-To-End Recognition. The competition received a total of 27 submissions over the different opened tasks. This report describes the datasets and the ground truth, details the performance evaluation protocols used and presents the final results along with a brief summary of the participating methods.
Original language | English |
---|---|
Journal | Proceedings of the International Conference on Document Analysis and Recognition, ICDAR |
Pages (from-to) | 1435-1443 |
Number of pages | 9 |
ISSN | 1520-5363 |
DOIs | |
Publication status | Published - 25 Jan 2018 |
Externally published | Yes |
Event | 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017 - Kyoto, Japan Duration: 9 Nov 2017 → 15 Nov 2017 |
Conference
Conference | 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017 |
---|---|
Country | Japan |
City | Kyoto |
Period | 09/11/2017 → 15/11/2017 |
Sponsor | et al., FxPaL, Glory, Hitachi, Media Drive, Sansan |
Bibliographical note
Funding Information:
*This work was supported by the Spanish project TIN2014-52072-P and the CERCA programme/Generalitat de Catalunya 1 R. Gomez is with the Computer Vision Center, Universitat Autonoma de Barcelona and Eurecat. 2 B. Shi is with the School of EIC, Huazhong University of Science and Technology 3,8 L. Gomez and D. Karatzas are with the Computer Vision Center, Universitat Autonoma de Barcelona. 4,6 L. Neumann and J. Matas are with the Center for Machine Perception, Czech Technical University. 5,7 A. Veit and S.Belongie are with the Cornell University and Cornell Tech.
Publisher Copyright:
© 2017 IEEE.
ID: 301826271