Introducing New AdaBoost Features for Real-Time Vehicle Detection

Abstract : This paper shows how to improve the real-time object detection in complex robotics applications, by exploring new visual features as AdaBoost weak classifiers. These new features are symmetric Haar filters (enforcing global horizontal and vertical symmetry) and N-connexity control points. Experimental evaluation on a car database show that the latter appear to provide the best results for the vehicle-detection problem.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [4 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00422587
Contributor : Fabien Moutarde <>
Submitted on : Wednesday, October 7, 2009 - 4:49:48 PM
Last modification on : Monday, November 12, 2018 - 11:04:23 AM
Document(s) archivé(s) le : Wednesday, June 16, 2010 - 12:28:41 AM

File

BSCogis2007paper_finalBis.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00422587, version 1
  • ARXIV : 0910.1293

Citation

Bogdan Stanciulescu, Amaury Breheret, Fabien Moutarde. Introducing New AdaBoost Features for Real-Time Vehicle Detection. COGIS'07 conference on COGnitive systems with Interactive Sensors, Nov 2007, Stanford, Palo Alto, United States. ⟨hal-00422587⟩

Share

Metrics

Record views

166

Files downloads

370