Comprehensive Parameter Sweep for Learning-based Detector on Traffic Lights

Publication: Research - peer-reviewArticle in proceeding

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.
Close

Details

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 languageEnglish
Title of host publicationAdvances in Visual Computing. ISVC 2016 : Lecture Notes in Computer Science (LNCS)
Volume10073
PublisherSpringer
Publication date10 Dec 2016
Pages92-100
DOI
StatePublished - 10 Dec 2016
Event - Las Vegas, United States

Conference

ConferenceISVC16: 12th International Symposium on Visual Computing
LocationMonte Carlo Resort & Casino
LandUnited States
ByLas Vegas
Periode12/12/201614/12/2016
Internetadresse

Download statistics

No data available
ID: 241387532