Learning the Distribution of Object Trajectories for Event Recognition

By Neil Johnson, David C. Hogg. 1995

How researchers see this paper

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

Citations in the corpus

An Incremental Self-Organizing Architecture for Sensorimotor Learning and Prediction (2017, background): “Johnson and Hogg [23] first proposed the use of multilayer self-organizing networks for the motion prediction of a tracked object.”
Trajectory Learning for Activity Understanding: Unsupervised, Multilevel, and Long-Term Adaptive Approach (2011, background): “Pioneering work by Johnson and Hogg [10] described outdoor motions with a flow vector and learned implicit temporal relationships using a leaky neural network.”
Learning object motion patterns for anomaly detection and improved object detection (2008, background): “Johnson et al. [9] presented a vector quantization based approach for learning typical trajectories of pedestrians in the scene, but they require entry/exit points to be marked manually.”
Intelligent multi-camera video surveillance: A review (2013, background): “In many surveillance systems (Johnson and Hogg, 1995; Stauffer and Grimson, 2000; Oliver et al., 2000; Haritaoglu et al., 2000; Brand and Kettnaker, 2000; Medioni et al., 2001; Honggeng and Nevatia, 2001; Hu et al., 2004; Wang et al., 2006; Morris and Trivedi, 2008; Wang et al., 2008, 2011),…”
Learning Variable-Length Markov Models of Behavior (2001, methodology): “In recent years, challenging problems such as human-computer interaction, automated visual surveillance and the realistic animation of human motion, have led to an increased interest in providing machines with the ability to learn and use models of human behaviour [15, 3, 13, 8].”
Learning Motion Patterns of People for Compliant Robot Motion (2005, background): “Johnson and Hogg [24] learn probability density functions of typical object trajectories to detect atypical behaviors.”
A system for learning statistical motion patterns (2006, methodology): “The first network and the input vector components to the second network are the same as in [20].”
Action recognition in videos acquired by a moving camera using motion decomposition of Lagrangian particle trajectories (2011, background): “Motion trajectories have been employed in a variety of problems for human action representation and recognition [1, 2, 3, 4, 13, 16, 17, 23].”
Trajectory-Based Anomalous Event Detection (2008, unknown): “The work by Johnson and Hogg [17] was probably one of the first researches in this direction , using vector quantization for the compact representation of trajectories and multilayer neural networks for the identification of common patterns.”
Robust Real-Time Unusual Event Detection using Multiple Fixed-Location Monitors (2008, result): “In contrast with most previous works reported, our algorithm satisfies all of these requirements (having of course its own limitations which will be discussed).”
Detecting unusual activity in video (2004, methodology): “In the second step the extracted features are used to develop models for the “normal” activities, either by hand or by applying supervised machine learning techniques [7].”
Dense Trajectories and Motion Boundary Descriptors for Action Recognition (2013, unknown): “Johnson and Hogg (1996) pro-pose to track humans and model the distribution of trajectories in order to identify atypical events.”
Learning semantic scene models from observing activity in visual surveillance (2005, methodology): “Johnson and Hogg [12] proposed a method for learning behavior models using a vector quantization method to learn typical routes taken by pedestrians from representative trajectories .”
Detecting anomalous events in videos by learning deep representations of appearance and motion (2017, background): “A pioneering work on this research line is [29], where object trajectories are modeled using probability density functions.”
Reachability Analysis and its Application to the Safety Assessment of Autonomous Cars (2010, background): “A similar work focuses more on the probability distribution of trajectories within a cluster [88].”
Learning Patterns of Activity Using Real-Time Tracking (2000, methodology): “Our method is most similar to the work of Johnson and Hogg [ 13 ].”
Learning activity patterns using fuzzy self-organizing neural network (2004, background): “Fernyhough et al. [22] establish spatio-temporal regions by learning the results of tracking the objects in a video sequence, and constructing a qualitative activity model by qualitative reasoning and statistical analysis.”
A survey on visual surveillance of object motion and behaviors (2004, background): “in [140] giving a more efficient prediction of object behaviors.”
Similarity based vehicle trajectory clustering and anomaly detection (2005, background): “Johnson et al. [1] used two competitive networks to learn the distribution of trajectory data.”
Anomaly detection in extremely crowded scenes using spatio-temporal motion pattern models (2009, methodology): “Trajectory-based approaches [7, 9, 13, 14], for example, track scene objects and describe the motion by their spatial location.”