A non-alternating graph hashing algorithm for large-scale image search
作者:
Highlights:
• A new relaxation for the graph hashing problem is presented.
• Unlike other solutions, no additional optimization variables added to the problem.
• Latent space transform, to significantly reduce memory demand during training.
• Proposed two optimization techniques to obtain binary codes.
• Achieving highly competitive performance at a much smaller time complexity.
摘要
•A new relaxation for the graph hashing problem is presented.•Unlike other solutions, no additional optimization variables added to the problem.•Latent space transform, to significantly reduce memory demand during training.•Proposed two optimization techniques to obtain binary codes.•Achieving highly competitive performance at a much smaller time complexity.
论文关键词:Hashing,Graph hashing,Image search,Spectral hashing
论文评审过程:Received 19 December 2020, Revised 28 April 2021, Accepted 20 March 2022, Available online 1 April 2022, Version of Record 9 April 2022.
论文官网地址:https://doi.org/10.1016/j.cviu.2022.103415