Nowadays, automotive companies are investing a relevant amount of resources for designing autonomous driving systems, driver assistance technologies and systems for assessing the driver’s attention. In this context, an important application consists of technologies for estimating the object distances in the scene, with a speciﬁc focus on pedestrians/cyclists. These technologies are usually based on LiDAR scanners, and thus require dedicated sensors and post-processing algorithms for estimating a depth map representing the distances between the vehicle and the surrounding objects. To obtain highly accurate distance estimations, methods based on Deep Learning (DL) and Convolutional Neural Networks (CNN) are being increasingly used for semantic segmentation in autonomous driving applications, considering either RGB images or LiDAR scans. In this paper, we propose the ﬁrst method in the literature able to estimate the distances of pedestrians from the vehicle by using only an RGB image and CNNs, without the need for any LiDAR scanner or any device designed for three-dimensional reconstruction of the scene. The proposed method is based on two CNNs: the ﬁrst one semantically segments the image regions representing pedestrians/cyclists, while the second one (DepthCNN) estimates a dense depth map of the scene. We evaluated our approach on a public dataset of RGB images and LiDAR scans captured in an automotive scenario, with results conﬁrming the feasibility of the proposed method.
Driver attention Assistance by Pedestrian/cyclist distance estimation from a single RGB Image: A CNN-based semantic segmentation approach / A. Genovese, V. Piuri, F. Rundo, F. Scotti, C. Spampinato - In: 2021 22nd IEEE International Conference on Industrial Technology (ICIT)[s.l] : IEEE, 2021. - ISBN 9781728157306. - pp. 875-880 (( Intervento presentato al 22. convegno IEEE Int. Conf. on Industrial Technology (ICIT) tenutosi a Valencia nel 2021 [10.1109/ICIT46573.2021.9453567].
|Titolo:||Driver attention Assistance by Pedestrian/cyclist distance estimation from a single RGB Image: A CNN-based semantic segmentation approach|
|Parole Chiave:||Deep Learning; CNN; Semantic Segmentation; Driver Attention; LiDAR|
|Settore Scientifico Disciplinare:||Settore INF/01 - Informatica|
Settore ING-INF/05 - Sistemi di Elaborazione delle Informazioni
|Progetto:||Multi-Owner data Sharing for Analytics and Integration respecting Confidentiality and Owner control (MOSAICrOWN)|
High quality Open data Publishing and Enrichment (HOPE)
|Data di pubblicazione:||2021|
|Digital Object Identifier (DOI):||http://dx.doi.org/10.1109/ICIT46573.2021.9453567|
|Tipologia:||Book Part (author)|
|Appare nelle tipologie:||03 - Contributo in volume|