Part-based Pedestrian Detection and Feature-based Tracking for Driver Assistance: Real-Time, Robust Algorithms and Evaluation

Antonio Prioletti, Andreas Møgelmose, Paolo Grislieri, Mohan Trivedi, Alberto Broggi, Thomas B. Moeslund

Research output: Contribution to journalJournal articleResearchpeer-review

80 Citations (Scopus)
695 Downloads (Pure)

Abstract

Detecting pedestrians is still a challenging task for automotive vision systems due to the extreme variability of targets, lighting conditions, occlusion, and high-speed vehicle motion. Much research has been focused on this problem in the last ten years and detectors based on classifiers have gained a special place among the different approaches presented. This paper presents a state-of-the-art pedestrian detection system based on a two-stage classifier. Candidates are extracted with a Haar cascade classifier trained with the Daimler Detection Benchmark data set and then validated through a part-based histogram-of-oriented-gradient (HOG) classifier with the aim of lowering the number of false positives. The surviving candidates are then filtered with feature-based tracking to enhance the recognition robustness and improve the results' stability. The system has been implemented on a prototype vehicle and offers high performance in terms of several metrics, such as detection rate, false positives per hour, and frame rate. The novelty of this system relies on the combination of a HOG part-based approach, tracking based on a specific optimized feature, and porting on a real prototype.
Original languageEnglish
JournalI E E E Transactions on Intelligent Transportation Systems
Volume14
Issue number3
Pages (from-to)1346-1359
Number of pages14
ISSN1524-9050
DOIs
Publication statusPublished - 2013

Fingerprint

Classifiers
Lighting
Detectors

Cite this

@article{6caf2ecc2ce840788088d5d87aa730c7,
title = "Part-based Pedestrian Detection and Feature-based Tracking for Driver Assistance: Real-Time, Robust Algorithms and Evaluation",
abstract = "Detecting pedestrians is still a challenging task for automotive vision systems due to the extreme variability of targets, lighting conditions, occlusion, and high-speed vehicle motion. Much research has been focused on this problem in the last ten years and detectors based on classifiers have gained a special place among the different approaches presented. This paper presents a state-of-the-art pedestrian detection system based on a two-stage classifier. Candidates are extracted with a Haar cascade classifier trained with the Daimler Detection Benchmark data set and then validated through a part-based histogram-of-oriented-gradient (HOG) classifier with the aim of lowering the number of false positives. The surviving candidates are then filtered with feature-based tracking to enhance the recognition robustness and improve the results' stability. The system has been implemented on a prototype vehicle and offers high performance in terms of several metrics, such as detection rate, false positives per hour, and frame rate. The novelty of this system relies on the combination of a HOG part-based approach, tracking based on a specific optimized feature, and porting on a real prototype.",
author = "Antonio Prioletti and Andreas M{\o}gelmose and Paolo Grislieri and Mohan Trivedi and Alberto Broggi and Moeslund, {Thomas B.}",
year = "2013",
doi = "10.1109/TITS.2013.2262045",
language = "English",
volume = "14",
pages = "1346--1359",
journal = "I E E E Transactions on Intelligent Transportation Systems",
issn = "1524-9050",
publisher = "IEEE",
number = "3",

}

Part-based Pedestrian Detection and Feature-based Tracking for Driver Assistance : Real-Time, Robust Algorithms and Evaluation. / Prioletti, Antonio; Møgelmose, Andreas; Grislieri, Paolo; Trivedi, Mohan; Broggi, Alberto; Moeslund, Thomas B.

In: I E E E Transactions on Intelligent Transportation Systems, Vol. 14, No. 3, 2013, p. 1346-1359.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Part-based Pedestrian Detection and Feature-based Tracking for Driver Assistance

T2 - Real-Time, Robust Algorithms and Evaluation

AU - Prioletti, Antonio

AU - Møgelmose, Andreas

AU - Grislieri, Paolo

AU - Trivedi, Mohan

AU - Broggi, Alberto

AU - Moeslund, Thomas B.

PY - 2013

Y1 - 2013

N2 - Detecting pedestrians is still a challenging task for automotive vision systems due to the extreme variability of targets, lighting conditions, occlusion, and high-speed vehicle motion. Much research has been focused on this problem in the last ten years and detectors based on classifiers have gained a special place among the different approaches presented. This paper presents a state-of-the-art pedestrian detection system based on a two-stage classifier. Candidates are extracted with a Haar cascade classifier trained with the Daimler Detection Benchmark data set and then validated through a part-based histogram-of-oriented-gradient (HOG) classifier with the aim of lowering the number of false positives. The surviving candidates are then filtered with feature-based tracking to enhance the recognition robustness and improve the results' stability. The system has been implemented on a prototype vehicle and offers high performance in terms of several metrics, such as detection rate, false positives per hour, and frame rate. The novelty of this system relies on the combination of a HOG part-based approach, tracking based on a specific optimized feature, and porting on a real prototype.

AB - Detecting pedestrians is still a challenging task for automotive vision systems due to the extreme variability of targets, lighting conditions, occlusion, and high-speed vehicle motion. Much research has been focused on this problem in the last ten years and detectors based on classifiers have gained a special place among the different approaches presented. This paper presents a state-of-the-art pedestrian detection system based on a two-stage classifier. Candidates are extracted with a Haar cascade classifier trained with the Daimler Detection Benchmark data set and then validated through a part-based histogram-of-oriented-gradient (HOG) classifier with the aim of lowering the number of false positives. The surviving candidates are then filtered with feature-based tracking to enhance the recognition robustness and improve the results' stability. The system has been implemented on a prototype vehicle and offers high performance in terms of several metrics, such as detection rate, false positives per hour, and frame rate. The novelty of this system relies on the combination of a HOG part-based approach, tracking based on a specific optimized feature, and porting on a real prototype.

U2 - 10.1109/TITS.2013.2262045

DO - 10.1109/TITS.2013.2262045

M3 - Journal article

VL - 14

SP - 1346

EP - 1359

JO - I E E E Transactions on Intelligent Transportation Systems

JF - I E E E Transactions on Intelligent Transportation Systems

SN - 1524-9050

IS - 3

ER -