A Real Time Object Detection Approach Applied to Reliable Pedestrian Detection

2007 ◽  
Author(s):  
Guanglin Ma ◽  
Su-Birm Park ◽  
Alexander Ioffe ◽  
Stefan Muller-Schneiders ◽  
Anton Kummert
Keyword(s):  
Object Detection ◽  
Real Time ◽  
Pedestrian Detection ◽  
Detection Approach

scholarly journals Object Detection Techniques in Videos

2019 ◽  
Vol 9 (2S) ◽  
pp. 616-620
Keyword(s):  
Anomaly Detection ◽  
Object Detection ◽  
Real Time ◽  
Pedestrian Detection ◽  
Research Attention ◽  
Detection Techniques ◽  
Future Work

Object detection in videos has increased its popularity because of its wider applications. It has gained more research attention now days as it is applicable in real time situations like pedestrian detection, anomaly detection, Self moving cars, sports, counting of people etc. This paper begins with the introduction of object detection and briefs the basic steps in the process. It also provides a review of various techniques and approaches used for object detection in videos. Discussion of every approach and limitations will provide several promising directions and guidelines for future work.

scholarly journals Real-time on-board pedestrian detection using generic single-stage algorithms and on-road databases

2020 ◽  
Vol 17 (5) ◽  
pp. 172988142092917
Author(s):  
Vicent Ortiz Castelló ◽  
Omar del Tejo Catalá ◽  
Ismael Salvador Igual ◽  
Juan-Carlos Perez-Cortes
Keyword(s):  
Object Detection ◽  
Real Time ◽  
Pedestrian Detection ◽  
Input Image ◽  
Performance Comparison ◽  
Detection Methods ◽  
Computational Time ◽  
Image Size ◽  
Single Class ◽  
Real Time System

Pedestrian detection is a particular case of object detection that helps to reduce accidents in advanced driver-assistance systems and autonomous vehicles. It is not an easy task because of the variability of the objects and the time constraints. A performance comparison of object detection methods, including both GPU and non-GPU implementations over a variety of on-road specific databases, is provided. Computer vision multi-class object detection can be integrated on sensor fusion modules where recall is preferred over precision. For this reason, ad hoc training with a single class for pedestrians has been performed and we achieved a significant increase in recall. Experiments have been carried out on several architectures and a special effort has been devoted to achieve a feasible computational time for a real-time system. Finally, an analysis of the input image size allows to fine-tune the model and get better results with practical costs.

Sign in / Sign up

Export Citation Format

Share Document