CV_Project

Deep Supervised Hashing for fast image retrieval

The objective of this project is to build a CNN based hashing method to convert image to a binary number for faster search. Keras framework is used with tensorflow as backend for this purpose.

Dataset

CIFAR - 10 https://www.cs.toronto.edu/~kriz/cifar.html

Code Overview

Python File	Description
Final_L1_1.py	Implementation using L1 distance and binary_crossentropy loss function
Final_L1_2.py	Implementation using L1 distance and Margin loss function (m=1) without regularizer
Final_L2_1.py	Implementation using L1 distance and Margin loss function(m-24) with L1 regularizer 0.01

References

• Haomiao Liu, Ruiping Wang, Shiguang Shan, Xilin Chen, "Deep Supervised Hashing for Fast Image Retrieval", CVPR 2016.