Learning Compact Binary Descriptors with Unsupervised Deep Neural Networks

• Published in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) pp. 1183 - 1192
• Main Authors: Lin, Kevin, Jiwen Lu, Chu-Song Chen, Jie Zhou
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2016
• Subjects: Binary codes; Machine learning; Neural networks; Optimization; Quantization (signal); Training data; Visualization
• ISSN: 1063-6919
• DOI: 10.1109/CVPR.2016.133

In this paper, we propose a new unsupervised deep learning approach called DeepBit to learn compact binary descriptor for efficient visual object matching. Unlike most existing binary descriptors which were designed with random projections or linear hash functions, we develop a deep neural network to learn binary descriptors in an unsupervised manner. We enforce three criterions on binary codes which are learned at the top layer of our network: 1) minimal loss quantization, 2) evenly distributed codes and 3) uncorrelated bits. Then, we learn the parameters of the networks with a back-propagation technique. Experimental results on three different visual analysis tasks including image matching, image retrieval, and object recognition clearly demonstrate the effectiveness of the proposed approach.