install following libraries: matplotlib, numpy, PIL, opencv, scikit-learn.
download haar-cascades for face recognition

Link to papers: https://www.cs.cmu.edu/~efros/courses/LBMV07/Papers/rubner-jcviu-00.pdf

http://vis.uky.edu/~cheung/courses/ee639_fall04/readings/ccv.pdf 

to make index execute:

	python index.py --index file-name.csv --dataset dresses/

Here file-name.csv is name of index file created and dresses is the directory where images are.

Making index may take upto 90 minutes for 2000 images.

to perform the search execute:

	python search.py --index file-name.csv --query image.jpg --result-path dresses/

here image.jpg is query image result-path is directory where images are and file-name.csv is the index file we created previously.

colordescriptor class helps to extract features and index in dataset.
searcher class helps in performing search.
index.py makes index
search.py performs search and retrieves results.

code is documented.

Thanks.