Sort and Deep-SORT Based Multi-Object Tracking for Mobile Robotics: Evaluation with New Data Association Metrics

Abstract

Multi-Object Tracking (MOT) techniques have been under continuous research and increasingly applied in a diverse range of tasks. One area in particular concerns its application in navigation tasks of assistive mobile robots, with the aim to increase the mobility and autonomy of people suffering from mobility decay, or severe motor impairments, due to muscular, neurological, or osteoarticular decay. Therefore, in this work, having in view navigation tasks for assistive mobile robots, an evaluation study of two MOTs by detection algorithms, SORT and Deep-SORT, is presented. To improve the data association of both methods, which are solved as a linear assignment problem with a generated cost matrix, a set of new object tracking data association cost matrices based on intersection over union, Euclidean distances, and bounding box metrics is proposed. For the evaluation of the MOT by detection in a real-time pipeline, the YOLOv3 is used to detect and classify the objects available on images. In addition, to perform the proposed evaluation aiming at assistive platforms, the ISR Tracking dataset, which represents the object conditions under which real robotic platforms may navigate, is presented. Experimental evaluations were also carried out on the MOT17 dataset. Promising results were achieved by the proposed object tracking data association cost matrices, showing an improvement in the majority of the MOT evaluation metrics compared to the default data association cost matrix. In addition, promising frame rate values were attained by the pipeline composed of the detector and the tracking module.

Towards a new intelligent traffic system based on deep learning and data integration

“…Utilizing you only look once (YOLOv4) software for real-time object detection system [29], as shown in Figure 11, we can detect cars with the best speed and accuracy. Utilizing deep SORT [30], the vehicles can be tracked. A sample of the result from the video processing is shown in Figure 12.…”

A Power-aware Vision-based Virtual Sensor for Real-Time Edge Computing

“…SORT, DeepSORT, and their variants tailored to improve the original algorithm have been evaluated both using MOT challenges metrics and other datasets. Most of the works compare SORT and DeepSORT against other state-ofthe-art trackers [15,24,31,34,40], sometimes also considering to skip some frames [23]. The same apply for IoU [8,26,36,39].…”

StereoYolo+DeepSORT: a framework to track fish from underwater stereo camera in situ

“…We used the latest version of Yolo (Yolov7) [14] for object detection as it is considered fast and can perform in real-time. The multiple objects tracking algorithm used in this framework is the deep SORT [15] because it has proved superior in tracking objects even when they get occluded or disappear from the scene for a short while. We adapted these methods with the focus on providing a 3D bounding box that defines the object, here the fish.…”