Strong-Background Restrained Cross Entropy Loss for Scene Text Detection
In this paper, we investigate the issue of class imbalance in scene text detection. Class Balanced Cross Entropy (CBCE) loss is often adopted for addressing this imbalance problem. We find that CBCE excessively restrains the backward gradients of background. Negative samples own extremely small weig...
Saved in:
Published in | 2019 International Joint Conference on Neural Networks (IJCNN) pp. 1 - 7 |
---|---|
Main Authors | , |
Format | Conference Proceeding |
Language | English |
Published |
IEEE
01.07.2019
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | In this paper, we investigate the issue of class imbalance in scene text detection. Class Balanced Cross Entropy (CBCE) loss is often adopted for addressing this imbalance problem. We find that CBCE excessively restrains the backward gradients of background. Negative samples own extremely small weights which are offered by CBCE during training of text detectors. These tiny weight values lead to insufficient learning of background. As a result, the CBCE-based text detection methods only can achieve sub-optimal performance. We propose a novel loss function, Strong-Background Restrained Cross Entropy (SBRCE), to deal with the disadvantage in CBCE. Specifically, SBRCE effectively down-weights the loss assigned to the strong background which means well-classified negative samples. Our SBRCE can make training focused on all positive samples and weak background(i.e., hard-classified negative samples). Moreover, it can prevent the enormous amount of strong background from overwhelming text detectors during training. Experimental results show that the proposed SBRCE can improve the performance of the efficient and accurate scene text detector (EAST) by F-score of 3.3% on ICDAR2015 dataset and 1.12% on MSRA-TD500 dataset, without sacrificing the training and testing speed of EAST. |
---|---|
AbstractList | In this paper, we investigate the issue of class imbalance in scene text detection. Class Balanced Cross Entropy (CBCE) loss is often adopted for addressing this imbalance problem. We find that CBCE excessively restrains the backward gradients of background. Negative samples own extremely small weights which are offered by CBCE during training of text detectors. These tiny weight values lead to insufficient learning of background. As a result, the CBCE-based text detection methods only can achieve sub-optimal performance. We propose a novel loss function, Strong-Background Restrained Cross Entropy (SBRCE), to deal with the disadvantage in CBCE. Specifically, SBRCE effectively down-weights the loss assigned to the strong background which means well-classified negative samples. Our SBRCE can make training focused on all positive samples and weak background(i.e., hard-classified negative samples). Moreover, it can prevent the enormous amount of strong background from overwhelming text detectors during training. Experimental results show that the proposed SBRCE can improve the performance of the efficient and accurate scene text detector (EAST) by F-score of 3.3% on ICDAR2015 dataset and 1.12% on MSRA-TD500 dataset, without sacrificing the training and testing speed of EAST. |
Author | Huang, Randong Xu, Bo |
Author_xml | – sequence: 1 givenname: Randong surname: Huang fullname: Huang, Randong email: huangrandong2015@ia.ac.cn organization: Institute of Automation, Chinese Academy of Sciences, Beijing, China – sequence: 2 givenname: Bo surname: Xu fullname: Xu, Bo email: xubo@ia.ac.cn organization: Institute of Automation, Chinese Academy of Sciences, Beijing, China |
BookMark | eNotj11LwzAYRqMouM39Ab3JH2jNm6Rpcql1ukmZ4Ob1aJO3o36kI4mw_Xsn7urhwOHAMyYXfvBIyA2wHICZu8VLtVzmnIHJtS44GH1GpqbUUHINRyrkORlxUJBJycorMo7xgzEujBEjMl-lMPht9tDYz20YfryjbxhTaHqPjlZhiJHO_NHZHWj9B90Q6MqiR7rGfaKPmNCmfvDX5LJrviJOTzsh70-zdTXP6tfnRXVfZz1nImVKGyaBo3NOuVajbK1TLZfSOSatlFzYzmpXKNXZppTYlU4oOLpgCmhbISbk9r_bI-JmF_rvJhw2p-PiFws9Tyo |
ContentType | Conference Proceeding |
DBID | 6IE 6IH CBEJK RIE RIO |
DOI | 10.1109/IJCNN.2019.8852198 |
DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan (POP) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Xplore / Electronic Library Online (IEL) IEEE Proceedings Order Plans (POP) 1998-present |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: RIE name: IEEE/IET Electronic Library url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Computer Science |
EISBN | 9781728119854 1728119855 |
EISSN | 2161-4407 |
EndPage | 7 |
ExternalDocumentID | 8852198 |
Genre | orig-research |
GroupedDBID | 29I 29O 6IE 6IF 6IH 6IK 6IL 6IM 6IN AAJGR ABLEC ADZIZ ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO IEGSK IJVOP IPLJI M43 OCL RIE RIL RIO RNS |
ID | FETCH-LOGICAL-i203t-6890412eddd6db8e4bcd6b244dd04c4423cfc8d566fca74ef7d361d6d1951bb33 |
IEDL.DBID | RIE |
IngestDate | Wed Jun 26 19:27:06 EDT 2024 |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-i203t-6890412eddd6db8e4bcd6b244dd04c4423cfc8d566fca74ef7d361d6d1951bb33 |
PageCount | 7 |
ParticipantIDs | ieee_primary_8852198 |
PublicationCentury | 2000 |
PublicationDate | 2019-July |
PublicationDateYYYYMMDD | 2019-07-01 |
PublicationDate_xml | – month: 07 year: 2019 text: 2019-July |
PublicationDecade | 2010 |
PublicationTitle | 2019 International Joint Conference on Neural Networks (IJCNN) |
PublicationTitleAbbrev | IJCNN |
PublicationYear | 2019 |
Publisher | IEEE |
Publisher_xml | – name: IEEE |
SSID | ssj0023993 |
Score | 1.7420852 |
Snippet | In this paper, we investigate the issue of class imbalance in scene text detection. Class Balanced Cross Entropy (CBCE) loss is often adopted for addressing... |
SourceID | ieee |
SourceType | Publisher |
StartPage | 1 |
SubjectTerms | Detectors Entropy Object detection Proposals Testing Training |
Title | Strong-Background Restrained Cross Entropy Loss for Scene Text Detection |
URI | https://ieeexplore.ieee.org/document/8852198 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1NT8JAEJ0AJ0-oYPzOHjxa6MfSbq8iBIkQo5BwI93dWWNICsFy0F_vzhYwGg_euk23bfbrvW7fmwG4MRxVKDLhaVK2c4G-l4mEwuMlYWz5cYcrMgqPxvFgyoezzqwCt3svDCI68Rm26ND9y9dLtaGtsrYQFmxSUYWq8MPSq7X_uCKg3Zli_LT9MOyOx6TcoqHgav1In-LQo1-H0e65pWhk0doUsqU-f4Vk_O-LHULz26fHnvYIdAQVzI-hvkvUwLbztgGDF9rwfvXuMrUgG0eu2TO-u-wQqFmXgJL1SLK--mCPVLBM1ta2yyCb2MWb3WPhFFt5E6b93qQ78LYpFLy30I8KLxYpBdRC7RJHCeRS6VhaSNfa54pbLqWMEtpyOqOyhKNJdBQH9trAMi8po-gEavkyx1NgacdOfpMaQ7AeaCEVqkRFQUy35NqcQYMaZr4qo2TMt21y_vfpCzigzimFr5dQK9YbvLLwXshr169fbcamkA |
link.rule.ids | 310,311,786,790,795,796,802,27956,55107 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1NT8JAEJ0gHvSECsZv9-DRQqFLu72KkILQGIWEG2l3Z40hKUTLQX-9O1vAaDx46zbdttmv97p9bwbgRnOULZEIR5GynQt0nUQEFB4vaPmGH7e5JKPwKPajCR9M29MS3G69MIhoxWdYp0P7L18t5Iq2yhpCGLAJxQ7sGpx3g8Kttf28Iqjd2GLcsNEfdOKYtFs0GGy9HwlULH70KjDaPLmQjczrqzyty89fQRn_-2oHUPt26rHHLQYdQgmzI6hsUjWw9cytQvRMW94vzl0i52TkyBR7wnebHwIV6xBUsi6J1pcfbEgFw2VNbbMQsrFZvtk95lazldVg0uuOO5GzTqLgvLZcL3d8EVJILVQ2dZRAnkrlpwbUlXK55IZNSS2FMqxOyyTgqAPl-U1zbdNwrzT1vGMoZ4sMT4CFbTP9dag1AXtTiVSiDKTX9OmWXOlTqFLDzJZFnIzZuk3O_j59DXvReDScDfvxwznsU0cVMtgLKOdvK7w0YJ-nV7aPvwBhcKnk |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2019+International+Joint+Conference+on+Neural+Networks+%28IJCNN%29&rft.atitle=Strong-Background+Restrained+Cross+Entropy+Loss+for+Scene+Text+Detection&rft.au=Huang%2C+Randong&rft.au=Xu%2C+Bo&rft.date=2019-07-01&rft.pub=IEEE&rft.eissn=2161-4407&rft.spage=1&rft.epage=7&rft_id=info:doi/10.1109%2FIJCNN.2019.8852198&rft.externalDocID=8852198 |