Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights

Morten Bornø Jensen; Mark Philip Philipsen; Mohan M. Trivedi; Thomas B. Moeslund

doi:10.1007/978-3-319-50832-0_10

Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights

Morten Bornø Jensen, Mark Philip Philipsen, Mohan M. Trivedi, Thomas B. Moeslund

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

1 Citation (Scopus)

417 Downloads (Pure)

Abstract

Determining the optimal parameters for a given detection algorithm is not straightforward and what ends up as the final values is mostly based on experience and heuristics. In this paper we investigate the influence of three basic parameters in the widely used Aggregate Channel Features (ACF) object detector applied for traffic light detec- tion. Additionally, we perform an exhaustive search for the optimal pa- rameters for the night time data from the LISA Traffic Light Dataset. The optimized detector reaches an Area-Under-Curve of 66.63 % on cal- culated precision-recall curve.

Original language	English
Title of host publication	Advances in Visual Computing. ISVC 2016 : Lecture Notes in Computer Science (LNCS)
Volume	10073
Publisher	Springer
Publication date	10 Dec 2016
Pages	92-100
DOIs	https://doi.org/10.1007/978-3-319-50832-0_10
Publication status	Published - 10 Dec 2016
Event	ISVC16: 12th International Symposium on Visual Computing - Monte Carlo Resort & Casino, Las Vegas, United States Duration: 12 Dec 2016 → 14 Dec 2016 http://www.isvc.net

Conference

Conference	ISVC16: 12th International Symposium on Visual Computing
Location	Monte Carlo Resort & Casino
Country/Territory	United States
City	Las Vegas
Period	12/12/2016 → 14/12/2016
Internet address	http://www.isvc.net

Access to Document

10.1007/978-3-319-50832-0_10

Jensen944Accepted author manuscript, 3.6 MBLicence: Unspecified

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{3c886d86884640d391cb9de394d65aaa,

title = "Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights",

abstract = "Determining the optimal parameters for a given detection algorithm is not straightforward and what ends up as the final values is mostly based on experience and heuristics. In this paper we investigate the influence of three basic parameters in the widely used Aggregate Channel Features (ACF) object detector applied for traffic light detec- tion. Additionally, we perform an exhaustive search for the optimal pa- rameters for the night time data from the LISA Traffic Light Dataset. The optimized detector reaches an Area-Under-Curve of 66.63 % on cal- culated precision-recall curve.",

author = "Jensen, {Morten Born{\o}} and Philipsen, {Mark Philip} and Trivedi, {Mohan M.} and Moeslund, {Thomas B.}",

year = "2016",

month = dec,

day = "10",

doi = "10.1007/978-3-319-50832-0_10",

language = "English",

volume = "10073",

pages = "92--100",

booktitle = "Advances in Visual Computing. ISVC 2016",

publisher = "Springer",

address = "Germany",

note = "ISVC16: 12th International Symposium on Visual Computing ; Conference date: 12-12-2016 Through 14-12-2016",

url = "http://www.isvc.net",

}

Jensen, MB, Philipsen, MP, Trivedi, MM & Moeslund, TB 2016, Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights. in Advances in Visual Computing. ISVC 2016: Lecture Notes in Computer Science (LNCS). vol. 10073, Springer, pp. 92-100, ISVC16: 12th International Symposium on Visual Computing, Las Vegas, United States, 12/12/2016. https://doi.org/10.1007/978-3-319-50832-0_10

Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights. / Jensen, Morten Bornø; Philipsen, Mark Philip; Trivedi, Mohan M. et al.
Advances in Visual Computing. ISVC 2016: Lecture Notes in Computer Science (LNCS). Vol. 10073 Springer, 2016. p. 92-100.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights

AU - Jensen, Morten Bornø

AU - Philipsen, Mark Philip

AU - Trivedi, Mohan M.

AU - Moeslund, Thomas B.

PY - 2016/12/10

Y1 - 2016/12/10

N2 - Determining the optimal parameters for a given detection algorithm is not straightforward and what ends up as the final values is mostly based on experience and heuristics. In this paper we investigate the influence of three basic parameters in the widely used Aggregate Channel Features (ACF) object detector applied for traffic light detec- tion. Additionally, we perform an exhaustive search for the optimal pa- rameters for the night time data from the LISA Traffic Light Dataset. The optimized detector reaches an Area-Under-Curve of 66.63 % on cal- culated precision-recall curve.

AB - Determining the optimal parameters for a given detection algorithm is not straightforward and what ends up as the final values is mostly based on experience and heuristics. In this paper we investigate the influence of three basic parameters in the widely used Aggregate Channel Features (ACF) object detector applied for traffic light detec- tion. Additionally, we perform an exhaustive search for the optimal pa- rameters for the night time data from the LISA Traffic Light Dataset. The optimized detector reaches an Area-Under-Curve of 66.63 % on cal- culated precision-recall curve.

U2 - 10.1007/978-3-319-50832-0_10

DO - 10.1007/978-3-319-50832-0_10

M3 - Article in proceeding

VL - 10073

SP - 92

EP - 100

BT - Advances in Visual Computing. ISVC 2016

PB - Springer

T2 - ISVC16: 12th International Symposium on Visual Computing

Y2 - 12 December 2016 through 14 December 2016

ER -

Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights

Abstract

Conference

Access to Document

AUB Link

Fingerprint

Cite this