Adaptive Hashing for Fast Similarity Search

With the staggering growth in image and video datasets, algorithms that provide fast similarity search and compact storage are crucial. Hashing methods that map the data into Hamming space have shown promise; however, many of these methods employ a batch-learning strategy in which the computational cost and memory requirements may become intractable and infeasible with larger and larger datasets. To overcome these challenges, we propose an online learning algorithm based on stochastic gradient descent in which the hash functions are updated iteratively with streaming data. In experiments with three image retrieval benchmarks, our online algorithm attains retrieval accuracy that is comparable to competing state-of-the-art batch-learning solutions, while our formulation is orders of magnitude faster and being online it is adaptable to the variations of the data. Moreover, our formulation yields improved retrieval performance over a recently reported online hashing technique, Online Kernel Hashing. (paper)

The source code is provided below:

The source code is implemented in MATLAB.  Please refer to the README file for details.  If you have any questions, please contact fcakir AT bu.edu.

If you download and use the code in research, we ask that you cite the below paper in the bibliography of your paper(s).

@inproceedings{cakir2015AdaptHash,
 title={Adaptive Hashing for Fast Similarity Search},
 author={Cakir, Fatih and Sclaroff, Stan},
 booktitle={IEEE International Conf. on Computer Vision (ICCV)},
 year={2015},
 organization={IEEE}
}

Code Changelog:

11.19.2015: Initial release.