Randomized dimensionality reduction of deep network features for image object recognition

Bui, H, Lech, M, Cheng, E, Neville, K, Wilkinson, R and Burnett, I 2018, 'Randomized dimensionality reduction of deep network features for image object recognition', in Vo Nguyen Quoc Bao and Tran Trung Duy (ed.) Proceedings of the 2nd International Conference on Recent Advances in Signal Processing, Telecommunications & Computing (SigTelCom 2018), Ho Chi Minh City, Vietnam, 29-31 January 2018, pp. 136-141.


Document type: Conference Paper
Collection: Conference Papers

Title Randomized dimensionality reduction of deep network features for image object recognition
Author(s) Bui, H
Lech, M
Cheng, E
Neville, K
Wilkinson, R
Burnett, I
Year 2018
Conference name SigTelCom 2018
Conference location Ho Chi Minh City, Vietnam
Conference dates 29-31 January 2018
Proceedings title Proceedings of the 2nd International Conference on Recent Advances in Signal Processing, Telecommunications & Computing (SigTelCom 2018)
Editor(s) Vo Nguyen Quoc Bao and Tran Trung Duy
Publisher IEEE
Place of publication United States
Start page 136
End page 141
Total pages 6
Abstract This study investigates data dimensionality reduction for image object recognition. The dimensionality reduction was applied to features extracted from an existing pre-trained Deep Neural Network (DNN) structure, the AlexNet. An analysis of the neurons in different layers of the AlexNet revealed an incremental increase in the pair-wise orthogonality between weight vectors of neurons in each layer, towards higher-level layers. This observation motivated the current study to evaluate the possibility of performing randomized dimensionality reduction by mimicking the observed orthogonality property of the high-level layers on activations of low-level layers of the AlexNet. Image object classification experiments have shown that the proposed random orthogonal projection method performed well in multiple tests, consistently outperforming the well-known statistics-based sparse random projection. Apart from being data independent, the proposed approach achieved performances comparable with the state-of-the-art techniques, but with lower computational requirements.
Subjects Signal Processing
Keyword(s) Feature extraction
Matrix decomposition
Dimensionality reduction
Sparse matrices
Principal component analysis
Biological neural networks
DOI - identifier 10.1109/SIGTELCOM.2018.8325778
Copyright notice © 2018 IEEE
ISBN 9781509006014
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 0 times in Scopus Article
Altmetric details:
Access Statistics: 15 Abstract Views  -  Detailed Statistics
Created: Thu, 31 Jan 2019, 11:26:00 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us