Adaboost with "Keypoint Presence Features" for Real-Time Vehicle Visual Detection

TL;DR

This paper introduces a real-time vehicle detection method using adaBoost with novel keypoint presence features, achieving high accuracy and revealing semantically meaningful keypoints.

Contribution

The paper proposes a new feature type for adaBoost that detects keypoint presence, enabling effective real-time vehicle detection with semantic insights.

Findings

95% recall and precision on test dataset

Keypoints correspond to meaningful vehicle parts

Effective real-time detection performance

Abstract

We present promising results for real-time vehicle visual detection, obtained with adaBoost using new original ?keypoints presence features?. These weak-classifiers produce a boolean response based on presence or absence in the tested image of a ?keypoint? (~ a SURF interest point) with a descriptor sufficiently similar (i.e. within a given distance) to a reference descriptor characterizing the feature. A first experiment was conducted on a public image dataset containing lateral-viewed cars, yielding 95% recall with 95% precision on test set. Moreover, analysis of the positions of adaBoost-selected keypoints show that they correspond to a specific part of the object category (such as ?wheel? or ?side skirt?) and thus have a ?semantic? meaning.