Profile Based Sub-Image Search in Image Databases

TL;DR

This paper introduces a profile-based feature vector for keypoints in images, enabling high-precision sub-image search without geometric validation, especially effective with small visual codebooks.

Contribution

The paper presents a novel profile feature for keypoints that improves sub-image search accuracy and efficiency without requiring geometric validation or large codebooks.

Findings

Achieves 81% precision on natural images

Outperforms traditional candidate generation by 31%

Effective with small codebooks of size 500

Abstract

Sub-image search with high accuracy in natural images still remains a challenging problem. This paper proposes a new feature vector called profile for a keypoint in a bag of visual words model of an image. The profile of a keypoint captures the spatial geometry of all the other keypoints in an image with respect to itself, and is very effective in discriminating true matches from false matches. Sub-image search using profiles is a single-phase process requiring no geometric validation, yields high precision on natural images, and works well on small visual codebook. The proposed search technique differs from traditional methods that first generate a set of candidates disregarding spatial information and then verify them geometrically. Conventional methods also use large codebooks. We achieve a precision of 81% on a combined data set of synthetic and real natural images using a codebook size of 500 for top-10 queries; that is 31% higher than the conventional candidate generation approach.