TY - JOUR
T1 - Bilinear Supervised Hashing Based on 2D Image Features
AU - Ding, Yujuan
AU - Wong, Wai Kueng
AU - Lai, Zhihui
AU - Zhang, Zheng
N1 - Funding Information:
Manuscript received August 9, 2018; revised November 12, 2018 and December 25, 2018; accepted December 25, 2018. Date of publication January 7, 2019; date of current version February 5, 2020. This work was supported by The Hong Kong Polytechnic University under Project Code RHQK. This paper was recommended by Associate Editor T. Mei. (Corresponding author: Wai Kueng Wong.) Y. Ding and W. K. Wong are with the Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hong Kong (e-mail: dingyujuan385@ gmail.com; calvin.wong@polyu.edu.hk).
Publisher Copyright:
© 1991-2012 IEEE.
PY - 2020/2
Y1 - 2020/2
N2 - Hashing has been recognized as an efficient representation learning method to effectively handle big data due to its low computational complexity and memory cost. Most of the existing hashing methods focus on learning the low-dimensional vectorized binary features based on the high-dimensional raw vectorized features. However, the studies on how to obtain preferable binary codes from the original 2D image features for retrieval is very limited. This paper proposes a bilinear supervised discrete hashing (BSDH) method based on 2D image features which utilizes bilinear projections to binarize the image matrix features such that the intrinsic characteristics in the 2D image space are preserved in the learned binary codes. Meanwhile, the bilinear projection approximation and vectorization binary codes regression are seamlessly integrated together to formulate the final robust learning framework. Furthermore, a discrete optimization strategy is developed to alternatively update each variable for obtaining the high-quality binary codes. In addition, two 2D image features, traditional SURF-based FVLAD feature, and CNN-based AlexConv5 feature are designed for further improving the performance of the proposed BSDH method. The results of extensive experiments conducted on four benchmark datasets show that the proposed BSDH method almost outperforms all competing hashing methods with different input features by different evaluation protocols.
AB - Hashing has been recognized as an efficient representation learning method to effectively handle big data due to its low computational complexity and memory cost. Most of the existing hashing methods focus on learning the low-dimensional vectorized binary features based on the high-dimensional raw vectorized features. However, the studies on how to obtain preferable binary codes from the original 2D image features for retrieval is very limited. This paper proposes a bilinear supervised discrete hashing (BSDH) method based on 2D image features which utilizes bilinear projections to binarize the image matrix features such that the intrinsic characteristics in the 2D image space are preserved in the learned binary codes. Meanwhile, the bilinear projection approximation and vectorization binary codes regression are seamlessly integrated together to formulate the final robust learning framework. Furthermore, a discrete optimization strategy is developed to alternatively update each variable for obtaining the high-quality binary codes. In addition, two 2D image features, traditional SURF-based FVLAD feature, and CNN-based AlexConv5 feature are designed for further improving the performance of the proposed BSDH method. The results of extensive experiments conducted on four benchmark datasets show that the proposed BSDH method almost outperforms all competing hashing methods with different input features by different evaluation protocols.
KW - 2D image feature
KW - Bilinear projection
KW - discrete optimization
KW - supervised hashing
UR - http://www.scopus.com/inward/record.url?scp=85059648422&partnerID=8YFLogxK
U2 - 10.1109/TCSVT.2019.2891246
DO - 10.1109/TCSVT.2019.2891246
M3 - Journal article
AN - SCOPUS:85059648422
SN - 1051-8215
VL - 30
SP - 590
EP - 602
JO - IEEE Transactions on Circuits and Systems for Video Technology
JF - IEEE Transactions on Circuits and Systems for Video Technology
IS - 2
M1 - 8604087
ER -