Face image retrieval based on shape and texture feature fusion
Abstract Humongous amounts of data bring various challenges to face image retrieval. This paper proposes an efficient method to solve those problems. Firstly, we use accurate facial landmark locations as shape features. Secondly, we utilise shape priors to provide discriminative texture features for...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2017-08-01
|
| Series: | Computational Visual Media |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1007/s41095-017-0091-7 |
| _version_ | 1819130732632604672 |
|---|---|
| author | Zongguang Lu Jing Yang Qingshan Liu |
| author_facet | Zongguang Lu Jing Yang Qingshan Liu |
| author_sort | Zongguang Lu |
| collection | DOAJ |
| description | Abstract Humongous amounts of data bring various challenges to face image retrieval. This paper proposes an efficient method to solve those problems. Firstly, we use accurate facial landmark locations as shape features. Secondly, we utilise shape priors to provide discriminative texture features for convolutional neural networks. These shape and texture features are fused to make the learned representation more robust. Finally, in order to increase efficiency, a coarse-tofine search mechanism is exploited to efficiently find similar objects. Extensive experiments on the CASIAWebFace, MSRA-CFW, and LFW datasets illustrate the superiority of our method. |
| first_indexed | 2024-12-22T09:04:17Z |
| format | Article |
| id | doaj.art-bf7f2ec633a14baa98a68c434230ef7c |
| institution | Directory Open Access Journal |
| issn | 2096-0433 2096-0662 |
| language | English |
| last_indexed | 2024-12-22T09:04:17Z |
| publishDate | 2017-08-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Computational Visual Media |
| spelling | doaj.art-bf7f2ec633a14baa98a68c434230ef7c2022-12-21T18:31:39ZengSpringerOpenComputational Visual Media2096-04332096-06622017-08-013435936810.1007/s41095-017-0091-7Face image retrieval based on shape and texture feature fusionZongguang Lu0Jing Yang1Qingshan Liu2School of Information and Control Engineering, Nanjing University of Information Science and TechnologySchool of Information and Control Engineering, Nanjing University of Information Science and TechnologySchool of Information and Control Engineering, Nanjing University of Information Science and TechnologyAbstract Humongous amounts of data bring various challenges to face image retrieval. This paper proposes an efficient method to solve those problems. Firstly, we use accurate facial landmark locations as shape features. Secondly, we utilise shape priors to provide discriminative texture features for convolutional neural networks. These shape and texture features are fused to make the learned representation more robust. Finally, in order to increase efficiency, a coarse-tofine search mechanism is exploited to efficiently find similar objects. Extensive experiments on the CASIAWebFace, MSRA-CFW, and LFW datasets illustrate the superiority of our method.http://link.springer.com/article/10.1007/s41095-017-0091-7face retrievalconvolutional neural networks (CNNs)coarse-to-fine |
| spellingShingle | Zongguang Lu Jing Yang Qingshan Liu Face image retrieval based on shape and texture feature fusion Computational Visual Media face retrieval convolutional neural networks (CNNs) coarse-to-fine |
| title | Face image retrieval based on shape and texture feature fusion |
| title_full | Face image retrieval based on shape and texture feature fusion |
| title_fullStr | Face image retrieval based on shape and texture feature fusion |
| title_full_unstemmed | Face image retrieval based on shape and texture feature fusion |
| title_short | Face image retrieval based on shape and texture feature fusion |
| title_sort | face image retrieval based on shape and texture feature fusion |
| topic | face retrieval convolutional neural networks (CNNs) coarse-to-fine |
| url | http://link.springer.com/article/10.1007/s41095-017-0091-7 |
| work_keys_str_mv | AT zongguanglu faceimageretrievalbasedonshapeandtexturefeaturefusion AT jingyang faceimageretrievalbasedonshapeandtexturefeaturefusion AT qingshanliu faceimageretrievalbasedonshapeandtexturefeaturefusion |