TY - JOUR
T1 - Unsupervised Local Feature Hashing for Image Similarity Search
AU - Liu, Li
AU - Yu, Mengyang
AU - Shao, Ling
PY - 2016/11
Y1 - 2016/11
N2 - The potential value of hashing techniques has led to it becoming one of the most active research areas in computer vision and multimedia. However, most existing hashing methods for image search and retrieval are based on global feature representations, which are susceptible to image variations such as viewpoint changes and background cluttering. Traditional global representations gather local features directly to output a single vector without the analysis of the intrinsic geometric property of local features. In this paper, we propose a novel unsupervised hashing method called unsupervised bilinear local hashing (UBLH) for projecting local feature descriptors from a high-dimensional feature space to a lower-dimensional Hamming space via compact bilinear projections rather than a single large projection matrix. UBLH takes the matrix expression of local features as input and preserves the feature-to-feature and image-to-image structures of local features simultaneously. Experimental results on challenging data sets including Caltech-256, SUN397, and Flickr 1M demonstrate the superiority of UBLH compared with state-of-the-art hashing methods.
AB - The potential value of hashing techniques has led to it becoming one of the most active research areas in computer vision and multimedia. However, most existing hashing methods for image search and retrieval are based on global feature representations, which are susceptible to image variations such as viewpoint changes and background cluttering. Traditional global representations gather local features directly to output a single vector without the analysis of the intrinsic geometric property of local features. In this paper, we propose a novel unsupervised hashing method called unsupervised bilinear local hashing (UBLH) for projecting local feature descriptors from a high-dimensional feature space to a lower-dimensional Hamming space via compact bilinear projections rather than a single large projection matrix. UBLH takes the matrix expression of local features as input and preserves the feature-to-feature and image-to-image structures of local features simultaneously. Experimental results on challenging data sets including Caltech-256, SUN397, and Flickr 1M demonstrate the superiority of UBLH compared with state-of-the-art hashing methods.
KW - Hashing
KW - image similarity search
KW - local feature
KW - unsupervised learning
U2 - 10.1109/TCYB.2015.2480966
DO - 10.1109/TCYB.2015.2480966
M3 - Article
VL - 46
SP - 2548
EP - 2558
JO - IEEE Transactions on Cybernetics
JF - IEEE Transactions on Cybernetics
SN - 2168-2267
IS - 11
ER -