Description

Human tracking in video sequences is a fundamental component of various applications and use cases, including human action recognition, attack and anomaly detection, suspect and criminal tracking, among others. In order to understand human actions in video sequences, it is necessary to first be able to track humans within them. Several works have been proposed for human tracking in video sequences, with applications either in real-time or disregarding the real-time criterion.

The principle of tracking a human in a video sequence is to locate the person in spatial and temporal dimensions, considering that video is a spatio-temporal modality. The major challenge in this research area is to find a compromise between response time and tracking accuracy.

The goal of this study is to implement an intelligent model for real-time tracking of humans in video sequences, using a hybridization of machine learning and interpolation techniques guided by bio-inspired algorithms. computer vision.

References:

[1] D. A. Maharani, C. Machbub, and L. Yulianti, “Real-time Human Tracking System using Histogram Intersection Distance in Firefly Optimization Based Particle Filter,” International Journal on Electrical Engineering and Informatics, vol. 13, no. 4, pp. 853–872, Dec. 2021, doi: 10.15676/ijeei.2021.13.4.7

[2] V.-T. Nguyen, A.-T. Nguyen, V.-T. Nguyen, and H.-A. Bui, “A Real-Time Human Tracking System Using Convolutional Neural Network and Particle Filter,” Intelligent Systems and Networks, vol. 243, pp. 411–417, 2021, doi: 10.1007/978-981-16-2094-2_50