A Robust Deep Networks based Multi-Object MultiCamera Tracking System for City Scale Traffic

TL;DR

This paper presents a deep learning-based multi-camera vehicle tracking system for city-scale traffic monitoring, combining Mask R-CNN, transfer learning, and Deep SORT to handle occlusions, illumination, and shadows effectively.

Contribution

It introduces an integrated framework utilizing Mask R-CNN, transfer learning, and Deep SORT for robust multi-camera vehicle tracking in urban environments.

Findings

Achieved an IDF1 score of 0.8289 on the AI City Challenge dataset.

Demonstrated high precision (0.9026) and recall (0.8527) in vehicle tracking.

Effectively handled occlusion, illumination variations, and shadows in city-scale traffic scenarios.

Abstract

Vision sensors are becoming more important in Intelligent Transportation Systems (ITS) for traffic monitoring, management, and optimization as the number of network cameras continues to rise. However, manual object tracking and matching across multiple non-overlapping cameras pose significant challenges in city-scale urban traffic scenarios. These challenges include handling diverse vehicle attributes, occlusions, illumination variations, shadows, and varying video resolutions. To address these issues, we propose an efficient and cost-effective deep learning-based framework for Multi-Object Multi-Camera Tracking (MO-MCT). The proposed framework utilizes Mask R-CNN for object detection and employs Non-Maximum Suppression (NMS) to select target objects from overlapping detections. Transfer learning is employed for re-identification, enabling the association and generation of vehicle tracklets across multiple cameras. Moreover, we leverage appropriate loss functions and distance measures to handle occlusion, illumination, and shadow challenges. The final solution identification module performs feature extraction using ResNet-152 coupled with Deep SORT based vehicle tracking. The proposed framework is evaluated on the 5th AI City Challenge dataset (Track 3), comprising 46 camera feeds. Among these 46 camera streams, 40 are used for model training and validation, while the remaining six are utilized for model testing. The proposed framework achieves competitive performance with an IDF1 score of 0.8289, and precision and recall scores of 0.9026 and 0.8527 respectively, demonstrating its effectiveness in robust and accurate vehicle tracking.