Title
Stereo-Based Framework for Pedestrian Detection with Partial Occlusion Handling
Abstract
The pedestrian detection literature has been recently extended by the availability of large-scale multisensory datasets, able to capture complementary aspects of the objects of interest, namely, appearance, motion, and depth. In this paper, we exploit this multimodal scenario to propose a new set of composite descriptors dubbed CO2, CO-variances of visual features and CO-occurrences of depth fields. Covariances of visual features allow us to integrate at low-level heterogeneous visual cues related to intensity and texture. Co-occurrences of depth fields are brand new descriptors, which use range information for characterizing the global shape of a pedestrian while being also able to identify its occluded parts. This paper illustrates how these descriptors can be instantiated and combined together, improving detection capabilities taking also benefit from the proper handling of occlusions. Experimental results show that CO2, fed into a standard discriminative classification system, set state-of-the-art performances on recent multi-modal intensity- and stereo-based pedestrian datasets.
Year
DOI
Venue
2012
10.1109/AVSS.2012.71
Advanced Video and Signal-Based Surveillance
Keywords
Field
DocType
pedestrian detection,composite descriptors,stereo-based framework,depth field,detection capability,brand new descriptors,partial occlusion handling,stereo-based pedestrian datasets,pedestrian detection literature,visual feature,new set,large-scale multisensory datasets,low-level heterogeneous visual cue,covariance matrix,image texture,computer vision,vectors,feature extraction,histograms,covariance,visualization,support vector machines,computer graphics,image classification
Histogram,Object detection,Computer vision,Pattern recognition,Image texture,Computer science,Visualization,Feature extraction,Artificial intelligence,Contextual image classification,Discriminative model,Pedestrian detection
Conference
ISBN
Citations 
PageRank 
978-1-4673-2499-1
0
0.34
References 
Authors
11
3
Name
Order
Citations
PageRank
Samuele Martelli1314.77
M. Cristani21928109.03
Vittorio Murino33277207.20