A method includes receiving an image 200 of a region near a vehicle and processing the image using a first neural network (NN) to determine one or more locations 202-208 where pedestrians are likely located within the image. The method also includes processing the one or more locations of the image using a second NN to determine that a pedestrian is present and notifying a driving assistance system or automated driving system that the pedestrian is present. Independent claims are also included for a system and a computer product. The first NN could generate a low-resolution saliency map 300, while the second NN processes locations at full resolution. The second NN could be trained using cropped ground truth bounding boxes.
Pedestrian detection with saliency maps
13.09.2017
Patent
Elektronische Ressource
Englisch
View-based object recognition using saliency maps
British Library Online Contents | 1999
|PEDESTRIAN DETECTION DEVICE, PEDESTRIAN DETECTION SYSTEM, AND PEDESTRIAN DETECTION METHOD
Europäisches Patentamt | 2018
|