Skip to main content
Kent Academic Repository

Compact Convolutional Neural Network Transfer Learning For Small-Scale Image Classification

Li, Zheng-xi, Song, Yan, McLoughlin, Ian Vince, Dai, Li-Rong (2016) Compact Convolutional Neural Network Transfer Learning For Small-Scale Image Classification. In: Acoustics, Speech and Signal Processing (ICASSP), 2016 IEEE International Conference on. Proceedings 2016 International Conference on Acoustics, Speech and Signal Processing. . IEEE E-ISBN 978-1-4799-9988-0. (doi:10.1109/ICASSP.2016.7472175) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided) (KAR id:55053)

The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided.
Official URL:
http://dx.doi.org/10.1109/ICASSP.2016.7472175

Abstract

Transfer learning methods have demonstrated state-of-the-art performance on various small-scale image classification tasks. This is generally achieved by exploiting the information from an ImageNet convolution neural network (ImageNet CNN). However, the transferred CNN model is generally with high computational complexity and storage requirement. It raises the issue for real-world applications, especially for some portable devices like phones and tablets without high-performance GPUs. Several approximation methods have been proposed to reduce the complexity by reconstructing the linear or non-linear filters (responses) in convolutional layers with a series of small ones., In this paper, we present a compact CNN transfer learning method for small-scale image classification. Specifically, it can be decomposed into fine-tuning and joint learning stages. In fine-tuning stage, a high-performance target CNN is trained by transferring information from the ImageNet CNN. In joint learning stage, a compact target CNN is optimized based on ground-truth labels, jointly with the predictions of the high-performance target CNN. The experimental results on CIFAR-10 and MIT Indoor Scene demonstrate the effectiveness and efficiency of our proposed method.

Item Type: Conference or workshop item (Paper)
DOI/Identification number: 10.1109/ICASSP.2016.7472175
Subjects: T Technology
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
Depositing User: Ian McLoughlin
Date Deposited: 19 Apr 2016 10:49 UTC
Last Modified: 17 Aug 2022 12:20 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/55053 (The current URI for this page, for reference purposes)

University of Kent Author Information

McLoughlin, Ian Vince.

Creator's ORCID: https://orcid.org/0000-0001-7111-2008
CReDIT Contributor Roles:
  • Depositors only (login required):

Total unique views for this document in KAR since July 2020. For more details click on the image.