HaoMood / crow

Python implementation of CroW for unsupervised image retrieval

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

     Cross-dimensional Weighting for Aggregated Deep Convolutional Features


DESCRIPTIONS
    This repository includes a Python implementation of CroW for unsupervised
    image retrieval. A full description of this work can be found in the paper.


REFERENCE
    Y, Kalantidis, et al. Cross-dimensional weighting for aggregated deep
    convolutional features. In Proceedings of the European Conference on
    Computer Vision Workshop, pages: 685--701, 2016.


PREREQUIREMENTS
    Caffe with Python interface supported
    Python2.7 with Numpy, PIL, sklearn supported


LAYOUT
    Data
    ./data/oxford5k/
        ./data/oxford5k/conv/            # Convolution feature maps
            ./data/oxford5k/conv/vgg16/all/
            ./data/oxford5k/conv/vgg16/crop/
            ./data/oxford5k/conv/vgg19/all/
            ./data/oxford5k/conv/vgg19/crop/
        ./data/oxford5k/groundtruth/     # 55*4=220 query groundtruth
        ./data/oxford5k/image/           # Raw .jpg images
            ./data/oxford5k/image/all/   # 5063 database images
            ./data/oxford5k/image/crop/  # 55 crop query images
    ./data/paris6k/
        # Similar to oxford5k, with 6392 database and 55 query images

    Documentations
    ./doc/                         # Automatically generated documents
    ./README                       # This file

    Library
    ./lib/                         # Third-party library files

    Source Code
    ./src/


USAGE
    Step 1. Extract crop query images and extract pool5 features.
        # Extract cropped query images.
        $ ./src/get_query.py --dataset oxford5k
        $ ./src/get_query.py --dataset paris6k

        # Extract pool5 features.
        $ ./src/get_conv.py --dataset oxford5k --model vgg19 --gpu 0
        $ ./src/get_conv.py --dataset paris6k --model vgg19 --gpu 0


    Step 2. Evaluate.
        $ ./src/evaluate.py --test oxford5k --via paris6k --model vgg19
        $ ./src/evaluate.py --test paris6k --via oxford5k --model vgg19


AUTHOR
    Hao Zhang: zhangh0214@gmail.com


COPYRIGHT
    2017 LAMDA


LICENSE
    CC BY-SA 3.0


NOTE
    20 images out of 6412 Paris6k images are broken. As a common practice,
    we manually removed them:
        paris_louvre_000136.jpg
        paris_louvre_000146.jpg
        paris_moulinrouge_000422.jpg
        paris_museedorsay_001059.jpg
        paris_notredame_000188.jpg
        paris_pantheon_000284.jpg
        paris_pantheon_000960.jpg
        paris_pantheon_000974.jpg
        paris_pompidou_000195.jpg
        paris_pompidou_000196.jpg
        paris_pompidou_000201.jpg
        paris_pompidou_000467.jpg
        paris_pompidou_000640.jpg
        paris_sacrecoeur_000299.jpg
        paris_sacrecoeur_000330.jpg
        paris_sacrecoeur_000353.jpg
        paris_triomphe_000662.jpg
        paris_triomphe_000833.jpg
        paris_triomphe_000863.jpg
        paris_triomphe_000867.jpg

About

Python implementation of CroW for unsupervised image retrieval


Languages

Language:Python 100.0%