Unsupervised discovery, modeling, and analysis of long term activities
ICVS'11 Proceedings of the 8th international conference on Computer vision systems
Event recognition in parking lot surveillance system
PRICAI'12 Proceedings of the 12th Pacific Rim international conference on Trends in Artificial Intelligence
Cross-domain traffic scene understanding by motion model transfer
Proceedings of the 4th ACM/IEEE international workshop on Analysis and retrieval of tracked events and motion in imagery stream
Traffic event classification at intersections based on the severity of abnormality
Machine Vision and Applications
Hi-index | 0.00 |
This paper presents a general framework for live video analysis. The activities of surveillance subjects are described using a spatio-temporal vocabulary learned from recurrent motion patterns. The repetitive nature of object trajectories is used to build a topographical scene description where nodes are points of interest (POI) and the edges correspond to activity paths (AP). The POI are learned through as a mixture of Gaussians and AP by clustering trajectories. The paths are probabilistically represented by hidden Markov models and adapt to temporal variations using online maximum likelihood regression (MLLR) and through a periodic batch update. Using the scene graph, new trajectories can be analyzed in online fashion to categorize past and present activity, predict future behavior, and detect abnormalities. The paths are first learned in unsupervised fashion by clustering trajectories and then probabilistically represented with hidden Markov models (HMM). Using the scene graph, new trajectories can be analyzed in online fashion to categorize past and present activity, predict future behavior, and detect abnormalities. The activity models adapt to scene variations, updating HMMs with maximum likelihood linear regression (MLLR) and adding abnormal activities given sufficient support. Experimental applications demonstrate the efficacy and generality of the proposed framework on an array of surveillance scenes.