Compression of Deep Neural Networks for Image Instance Retrieval
Name
1701.04923.pdf
Size
614.33 KB
Format
Adobe PDF
Checksum (MD5)
bcd0eccb1f3e0a3140c0d6ac68c3e895
Author(s)
Chandrasekhar, Vijay
Lin, Jie
Morere, Olivier
Veillard, Antoine
Duan, Lingyu
Liao, Qianli
Poggio, Tomaso A
Date Issued
May 2017
Journal
2017 Data Compression Conference (DCC)
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Citation
Chandrasekhar, Vijay et al. “Compression of Deep Neural Networks for Image Instance Retrieval.” 2017 Data Compression Conference (DCC), April 4-7 2017, Snowbird, Utah, USA, Institute of Electrical and Electronics Engineers (IEEE), May 2017 © 2017 Institute of Electrical and Electronics Engineers (IEEE)
Version
Author's final manuscript
Abstract
Image instance retrieval is the problem of retrieving images from a database which contain the same object. Convolutional Neural Network (CNN) based descriptors are becoming the dominant approach for generating global image descriptors for the instance retrieval problem. One major drawback of CNN-based global descriptors is that uncompressed deep neural network models require hundreds of megabytes of storage making them inconvenient to deploy in mobile applications or in custom hardware. In this work, we study the problem of neural network model compression focusing on the image instance retrieval task. We study quantization, coding, pruning and weight sharing techniques for reducing model size for the instance retrieval problem. We provide extensive experimental results on the trade-off between retrieval performance and model size for different types of networks on several data sets providing the most comprehensive study on this topic. We compress models to the order of a few MBs: Two orders of magnitude smaller than the uncompressed models while achieving negligible loss in retrieval performance1.
MIT Department
Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
McGovern Institute for Brain Research at MIT
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1109/DCC.2017.93
