szupzp/vgg16-image-retrieval

Uses TensorFlow and FC2 features to match test images to the same category given a query image as input

VGG16-Image-Retrieval

Uses TensorFlow and FC2 features to match test images to the same category given a query image as input

How to use:

• Install TensorFlow by following instructions here: https://www.tensorflow.org/versions/r0.10/get_started/os_setup.html
• Install python pre-reqs
  • sudo pip install numpy
  • sudo pip install scipy
• Uncompress the dataset folder
• Simply execute the vgg_example.py file
  • python vgg16_example.py

Once it is up and running, every so often you will see prints similar to the following:

Current Correct list: ['image_00748.jpg', 'image_00749.jpg', 'image_00750.jpg', 'image_00751.jpg']

Matches are:

• Distance: 2.45867538452, FileName: image_00748.jpg
• Distance: 22.8059749603, FileName: image_00751.jpg
• Distance: 37.0147171021, FileName: image_00750.jpg
• Distance: 38.9278831482, FileName: image_00749.jpg
• Precision@4: 1.0

Notes:

• If you are having issues with the dataset .zip file please try downloading from here: https://drive.google.com/file/d/0B5DVPd_zGgc8Vm9HRVFkSmpCT1k/view?usp=sharing

This shows the similarity distance between the query and the matches as well as the filename of the matched images. This also shows the precision@4 compared to the ground truth. A file containing these prints as well as final results will be created as "Last_Run.txt"