An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework
This paper proposes an efficient technique for learning a discriminative codebook for scene categorization. A state-of-the-art approach for scene categorization is the Bag-of-Words (BoW) framework, where codebook generation plays an important role in determining the performance of the system. Tradit...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2016
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/81682 http://hdl.handle.net/10220/40930 |
| _version_ | 1826123503544303616 |
|---|---|
| author | Li, Zhen Yap, Kim-Hui |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Li, Zhen Yap, Kim-Hui |
| author_sort | Li, Zhen |
| collection | NTU |
| description | This paper proposes an efficient technique for learning a discriminative codebook for scene categorization. A state-of-the-art approach for scene categorization is the Bag-of-Words (BoW) framework, where codebook generation plays an important role in determining the performance of the system. Traditionally, the codebook generation methods adopted in the BoW techniques are designed to minimize the quantization error, rather than optimize the classification accuracy. In view of this, this paper tries to address the issue by careful design of the codewords such that the resulting image histograms for each category will retain strong discriminating power, while the online categorization of the testing image is as efficient as in the baseline BoW. The codewords are refined iteratively to improve their discriminative power offline. The proposed method is validated on UIUC Scene-15 dataset and NTU Scene-25 dataset and it is shown to outperform other state-of-the-art codebook generation methods in scene categorization. |
| first_indexed | 2024-10-01T06:05:25Z |
| format | Journal Article |
| id | ntu-10356/81682 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:05:25Z |
| publishDate | 2016 |
| record_format | dspace |
| spelling | ntu-10356/816822020-03-07T13:56:08Z An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework Li, Zhen Yap, Kim-Hui School of Electrical and Electronic Engineering Scene categorization Codebook learning This paper proposes an efficient technique for learning a discriminative codebook for scene categorization. A state-of-the-art approach for scene categorization is the Bag-of-Words (BoW) framework, where codebook generation plays an important role in determining the performance of the system. Traditionally, the codebook generation methods adopted in the BoW techniques are designed to minimize the quantization error, rather than optimize the classification accuracy. In view of this, this paper tries to address the issue by careful design of the codewords such that the resulting image histograms for each category will retain strong discriminating power, while the online categorization of the testing image is as efficient as in the baseline BoW. The codewords are refined iteratively to improve their discriminative power offline. The proposed method is validated on UIUC Scene-15 dataset and NTU Scene-25 dataset and it is shown to outperform other state-of-the-art codebook generation methods in scene categorization. ASTAR (Agency for Sci., Tech. and Research, S’pore) 2016-07-13T08:12:58Z 2019-12-06T14:36:01Z 2016-07-13T08:12:58Z 2019-12-06T14:36:01Z 2013 Journal Article Li, Z., & Yap, K.-H. (2013). An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework. Image and Vision Computing, 31(10), 748-755. 0262-8856 https://hdl.handle.net/10356/81682 http://hdl.handle.net/10220/40930 10.1016/j.imavis.2013.07.001 en Image and Vision Computing © 2013 Elsevier. |
| spellingShingle | Scene categorization Codebook learning Li, Zhen Yap, Kim-Hui An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework |
| title | An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework |
| title_full | An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework |
| title_fullStr | An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework |
| title_full_unstemmed | An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework |
| title_short | An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework |
| title_sort | efficient approach for scene categorization based on discriminative codebook learning in bag of words framework |
| topic | Scene categorization Codebook learning |
| url | https://hdl.handle.net/10356/81682 http://hdl.handle.net/10220/40930 |
| work_keys_str_mv | AT lizhen anefficientapproachforscenecategorizationbasedondiscriminativecodebooklearninginbagofwordsframework AT yapkimhui anefficientapproachforscenecategorizationbasedondiscriminativecodebooklearninginbagofwordsframework AT lizhen efficientapproachforscenecategorizationbasedondiscriminativecodebooklearninginbagofwordsframework AT yapkimhui efficientapproachforscenecategorizationbasedondiscriminativecodebooklearninginbagofwordsframework |