Learning semantic scene models from observing activity in visual surveillance

By Dimitrios Makris, T. Ellis. IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics), 2005

305 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/110d90353f33ba88d40a40006189e7b1f52e8d57

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Comparative study of background subtraction algorithms (2010, background): “In this perspective, motion detection is often the first step of a multi-stage computer vision system [8] [20] [24] [25] (car tracking, person recognition, wild-life monitoring, etc.).”
Learning object motion patterns for anomaly detection and improved object detection (2008, background): “Makris et al. [12] have presented a technique in which different regions of the scene are labelled as entry/exit zones, junctions, paths and stop zones.”
A Survey of Vision-Based Trajectory Learning and Analysis for Surveillance (2008, methodology): “We presented a few online adaptive techniques [23], [67] but all path learning methods should be able to introduce new models”
Vessel Pattern Knowledge Discovery from AIS Data: A Framework for Anomaly Detection and Route Prediction (2013, background): “, [16]), entry and exit points are related to the”
Detecting anomalous events in videos by learning deep representations of appearance and motion (2017, background): “Markris and Ellis [31] proposed a Bayesian approach for detecting abnormal trajectories based on annotated scene semantics.”
Survey on Contemporary Remote Surveillance Systems for Public Safety (2010, background): “In some cases, surveillance performs persuasion, discouraging unacceptable behavior that can no longer be performed anonymously, recording and logging events for evidential reasons, or offering remote observation of sensitive locations where access control is crucial [19].”
Trajectory Learning for Activity Understanding: Unsupervised, Multilevel, and Long-Term Adaptive Approach (2011, methodology): “The POI are learned through a 2D mixture of Gaussian zone modeling procedure [17].”
Understanding collective crowd behaviors: Learning a Mixture model of Dynamic pedestrian-Agents (2012, background): “Some approaches learned the motion patterns through clustering trajectories [11, 21, 22], and faced the challenge of fragmentation of trajectories in crowded scenes.”
Trajectory-Based Anomalous Event Detection (2008, methodology): “Makris and Ellis [20], [21] proposed a method for labelling the scene with topological information which is then used within a Bayesian approach to detect anomalous trajectories.”
A system for learning statistical motion patterns (2006, methodology): “In contrast with Johnson and Hogg [20], Sumpter and Bulpitt [43], Makris and Ellis [25], and Junejo et al. [22], our work has introduced probability distributions to the motion patterns; the number of motion patterns is evaluated; and anomaly detection and behavior prediction are implemented based…”