Contextual Combination of Appearance and Motion for Intersection Videos with Vehicles and Pedestrians

Abstract

Abstract. Object detection and classification is challenging problem for vision-based intersection monitoring since traditional motion-based techniques work poorly when pedestrians or vehicles stop due to traffic signals. In this work, we present a method for vehicle and pedestrian recognition at intersections that benefits from both motion and appearance cues in video surveillance. Vehicle and pedestrian recognition performance is compared using motion, appearance and combined cues in contextually relevant stop areas to improve recognition. Experimental evaluation shows 5% average improvement for vehicle and pedestrian recognition at two Las Vegas intersections.

Vision-based pedestrian behavior analysis at intersections

“…However, it fails in familiar intersection situations. 11 1. Although motion is widely used for highway scenarios, it is not consistent at intersections since traffic signals force pedestrians to stop.…”

“…However, we previously demonstrated high performance for appearance-based pedestrian detection techniques in specially defined mix areas (i.e., around signals and cross-walks). 11 Two famous features in computer vision, Haar-like features and local binary patterns (LBP), are evaluated in the system. Viola and Jones 13 used the Haar-like features, which are adjacent rectangular regions in a small window that represent intensity patterns, along with a cascaded Adaboost classifier for real-time detection of rigid objects.…”

“…The tracking system benefits from some improvements at detection and tracking steps through using appearance-based classifiers, which do not rely on motion. 11 In order to choose an appropriate classifier, famous available datasets are trained and evaluated separately, and UNLV dataset is developed from the available surveillance cameras. The rest of the paper explains the system and the major contributions with more details.…”

Structured behaviour prediction of on‐road vehicles via deep forest

“…one forest tree 1 tree n Fig. 4 Example of random forest based on structured labels Experiments and analysis: For collecting the label with structured information, MIT-CBCL Car Database [14] is employed. The MIT traffic dataset is for research on activity analysis and crowded scenes.…”

“…For collecting the label with structured information, MIT‐CBCL Car Database [14] is employed. The MIT traffic dataset is for research on activity analysis and crowded scenes.…”