Awesome

Intro

This repository contains a CBIR (content-based image retrieval) system

Extract query image's feature, and retrieve similar ones from image database

<h6><a href='https://winstonhsu.info/2017f-mmai/'>Image src</a></h6>

Part1: Feature Extraction

In this system, I implement several popular image features:

color-based
- RGB histogram
texture-based
- gabor filter
shape-based
deep methods
- VGG net
- Residual net

all features are modulized

Feature Fusion

Some features are not robust enough, and turn to feature fusion

fusion.py

Dimension Reduction

The curse of dimensionality told that vectors in high dimension will sometimes lose distance property

Random Projection

Part2: Evaluation

CBIR system retrieves images based on feature similarity

Robustness of system is evaluated by MMAP (mean MAP), the evaluation formulas is refer to <a href='http://web.stanford.edu/class/cs276/handouts/EvaluationNew-handout-1-per.pdf' target="_blank">here</a>

image AP : average of precision at each hit
- depth=K means the system will return top-K images
- a correct image in top-K is called a hit
- AP = (hit1.precision + hit2.precision + ... + hitH.precision) / H
class1 MAP = (class1.img1.AP + class1.img2.AP + ... + class1.imgM.AP) / M
MMAP = (class1.MAP + class2.MAP + ... + classN.MAP) / N

Implementation of evaluation can found at evaluate.py

My database contains 25 classes, each class with 20 images, 500 images in total, depth=K will return top-K images from database

Method	color	daisy	edge	gabor	HOG	vgg19	resnet152
Mean MAP (depth=10)	0.614	0.468	0.301	0.346	0.450	0.914	0.944

Part3: Image Retrieval (return top 5 of each method)

Let me show some results of the system

query1 - women dress

query <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1.jpg' padding='5px' height="80px"></img>

color <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-color.jpg' padding='5px' height="80px"></img>

daisy <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-daisy.jpg' padding='5px' height="80px"></img>

edge <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-edge.jpg' padding='5px' height="80px"></img>

gabor <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-gabor.jpg' padding='5px' height="80px"></img>

HOG <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-hog.jpg' padding='5px' height="80px"></img>

VGG19 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-vgg.jpg' padding='5px' height="80px"></img>

Resnet152 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query1-women_dress/query1-res.jpg' padding='5px' height="80px"></img>

query2 - orange

query <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2.jpg' padding='5px' height="80px"></img>

color <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-color.jpg' padding='5px' height="80px"></img>

daisy <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-daisy.jpg' padding='5px' height="80px"></img>

edge <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-edge.jpg' padding='5px' height="80px"></img>

gabor <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-gabor.jpg' padding='5px' height="80px"></img>

HOG <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-hog.jpg' padding='5px' height="80px"></img>

VGG19 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-vgg.jpg' padding='5px' height="80px"></img>

Resnet152 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query2-orange/query2-res.jpg' padding='5px' height="80px"></img>

query3 - NBA jersey

query <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3.jpg' padding='5px' height="80px"></img>

color <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-color.jpg' padding='5px' height="80px"></img>

daisy <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-daisy.jpg' padding='5px' height="80px"></img>

edge <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-edge.jpg' padding='5px' height="80px"></img>

gabor <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-gabor.jpg' padding='5px' height="80px"></img>

HOG <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-hog.jpg' padding='5px' height="80px"></img>

VGG19 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-vgg.jpg' padding='5px' height="80px"></img>

Resnet152 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query3-nba_jersey/query3-res.jpg' padding='5px' height="80px"></img>

query4 - snack

query <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4.jpg' padding='5px' height="80px"></img>

color <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-color.jpg' padding='5px' height="80px"></img>

daisy <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-daisy.jpg' padding='5px' height="80px"></img>

edge <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-edge.jpg' padding='5px' height="80px"></img>

gabor <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-gabor.jpg' padding='5px' height="80px"></img>

HOG <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-hog.jpg' padding='5px' height="80px"></img>

VGG19 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-vgg.jpg' padding='5px' height="80px"></img>

Resnet152 <img align='center' style="border-color:gray;border-width:2px;border-style:dashed" src='result/retrieval_result/query4-snack/query4-res.jpg' padding='5px' height="80px"></img>

Part4: Usage of Repository

If you are interesting with the results, and want to try your own images,

Please refer to USAGE.md

The details are written inside.

Author

Po-Chih Huang / @pochih