Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 14, 2026

Utilizing vmTracking to Improve the Accuracy of Multi-Animal Pose Estimation in Rodent Social Behavior Studies

Utilizing vmTracking to Improve the Accuracy of Multi-Animal Pose Estimation in Rodent Social Behavior Studies

Published on: November 7, 2025

Object trajectory-based activity classification and recognition using hidden Markov models.

Faisal I Bashir¹, Ashfaq A Khokhar, Dan Schonfeld

¹Retica Systems, Inc., Waltham, MA 01801, USA. fbashir@gmail.com

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society

|July 4, 2007

Summary

This study introduces a new method for object activity recognition using motion trajectories. Hidden Markov Models (HMMs) with Principal Component Analysis (PCA) coefficients significantly improve classification accuracy compared to other techniques.

Related Concept Videos

Orthogonal Trajectories

Orthogonal Trajectories

Orthogonal trajectories describe the geometric relationship between two families of curves that intersect each other at right angles. One illustrative case involves a family of parabolas that open sideways along the x-axis. These curves share a common shape but differ by a scaling parameter, resulting in a set of curves that all pass through the origin and widen at different rates.Determining Orthogonal TrajectoriesTo identify the orthogonal trajectories for these parabolas, the first step...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A Deep Learning-Based Approach to Characterize Skull Physical Properties: A Phantom Study.

Journal of biophotonics·2024

Same author

Deep Learning Analyses of Brain MRI to Identify Sustained Attention Deficit in Treated Obstructive Sleep Apnea: A Pilot Study.

Sleep and vigilance·2022

Same author

Inferring excitation-inhibition dynamics using a maximum entropy model unifying brain structure and function.

Network neuroscience (Cambridge, Mass.)·2022

Same author

Autocatalytic-protection for an unknown locus CRISPR-Cas countermeasure for undesired mutagenic chain reactions.

Journal of theoretical biology·2021

Same author

Connectome Signatures of Hyperexcitation in Cognitively Intact Middle-Aged Female APOE-ε4 Carriers.

Cerebral cortex (New York, N.Y. : 1991)·2020

Same author

Deep learning applied to polysomnography to predict blood pressure in obstructive sleep apnea and obesity hypoventilation: a proof-of-concept study.

Journal of clinical sleep medicine : JCSM : official publication of the American Academy of Sleep Medicine·2020

Area of Science:

Computer Vision
Machine Learning
Pattern Recognition

Background:

Motion trajectories contain valuable spatiotemporal data for activity recognition.
Existing methods using Gaussian Mixture Models (GMMs) struggle to capture temporal dynamics.

Purpose of the Study:

To develop novel classification algorithms for object activity recognition from motion trajectories.
To enhance temporal relationship modeling in trajectory-based activity recognition.

Main Methods:

Trajectory segmentation at curvature change points.
Representation of subtrajectories using Principal Component Analysis (PCA) coefficients.
Application of Hidden Markov Models (HMMs) for temporal modeling, with data-driven design for states and topology.

Related Experiment Videos

Last Updated: Jul 14, 2026

Utilizing vmTracking to Improve the Accuracy of Multi-Animal Pose Estimation in Rodent Social Behavior Studies

Utilizing vmTracking to Improve the Accuracy of Multi-Animal Pose Estimation in Rodent Social Behavior Studies

Published on: November 7, 2025

Main Results:

Gaussian Mixture Models (GMMs) alone are insufficient for capturing temporal ordering.
The proposed HMM-based scheme using PCA coefficients demonstrates superior performance.
Experiments on a large dataset (5700+ trajectories, 85 classes) validate the approach.

Conclusions:

HMMs effectively model temporal relations in object motion trajectories.
The combination of PCA and HMMs offers a robust and accurate solution for activity recognition.
This method outperforms existing trajectory classification techniques.