Paper Info
Title
Deep hashing with multi-task learning for large-scale instance-level vehicle search
Abstract
Hashing is a hot research topic in large-scale image search, due to its low memory cost and fast search speed. Recently, deep hashing, which adapts deep convolutional neural networks into hashing, has attracted much attention. In this paper, we propose a new supervised deep hashing method to deal with large-scale instance-level vehicle search, and make the following contributions. Firstly, multi-task learning is employed to learn the hash code, which exploits the available multiple labels of each vehicle, i.e., ID, model, and color. Secondly, differing from several deep hashing methods, which utilize sigmoid or tanh as the activation function of the hash layer, rectified linear unit is adopted in this paper and shows better performance. Thirdly, taking GoogLeNet as the base network, we show that search performance can be promoted significantly, by learning the network's parameters from scratch on our vehicle data. Finally, we perform extensive experiments on a large-scale dataset with up to one million vehicles. The experimental results demonstrate the effectiveness of the proposed method, which outperforms single task deep hashing methods with classification and triplet ranking losses, respectively.
Year
DOI
Venue
2017
10.1109/ICMEW.2017.8026274
2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)
Keywords
Field
DocType
Hashing,Deep Learning,Multi-Task Learning,Vehicle Search,Large Scale
Locality-sensitive hashing,Data mining,Hopscotch hashing,Double hashing,Computer science,Universal hashing,Feature hashing,Artificial intelligence,Dynamic perfect hashing,Machine learning,Hash table,Open addressing
Conference
ISSN
ISBN
Citations 
2330-7927
978-1-5386-0561-5
0
PageRank 
References 
Authors
0.34
20
7
Name
Order
Citations
PageRank
Dawei Liang1171.31
Yan Ke22581191.93
Yaowei Wang313429.62
Wei Zeng4111.61
Qingsheng Yuan541.73
Xiuguo Bao601.01
Yonghong Tian71057102.81