Multi-Class Multi-Object Tracking using Changing Point Detection

TL;DR

This paper introduces a multi-class multi-object tracking method using Bayesian filtering that combines CNN-based detection, KLT motion detection, and change point detection to improve robustness against drift and occlusion in challenging videos.

Contribution

It proposes a novel framework integrating change point detection with Bayesian filtering for multi-class multi-object tracking, handling abrupt changes and occlusions effectively.

Findings

Outperforms state-of-the-art methods on ImageNet VID and MOT benchmarks.

Effectively detects abrupt changes and occlusions in tracking.

Demonstrates robustness across multiple object classes.

Abstract

This paper presents a robust multi-class multi-object tracking (MCMOT) formulated by a Bayesian filtering framework. Multi-object tracking for unlimited object classes is conducted by combining detection responses and changing point detection (CPD) algorithm. The CPD model is used to observe abrupt or abnormal changes due to a drift and an occlusion based spatiotemporal characteristics of track states. The ensemble of convolutional neural network (CNN) based object detector and Lucas-Kanede Tracker (KLT) based motion detector is employed to compute the likelihoods of foreground regions as the detection responses of different object classes. Extensive experiments are performed using lately introduced challenging benchmark videos; ImageNet VID and MOT benchmark dataset. The comparison to state-of-the-art video tracking techniques shows very encouraging results.