The advancement of multi-object tracking (MOT) technologies presents the dual challenge of maintaining high performance while addressing critical security and privacy concerns. In applications such as pedestrian tracking, where sensitive personal data is involved, the potential for privacy violations and data misuse becomes a significant issue if data is transmitted to external servers. To mitigate these risks, processing data directly on an edge device, such as a smart camera, has emerged as a viable solution. Edge computing ensures that sensitive information remains local, thereby aligning with stringent privacy principles and significantly reducing network latency. However, the implementation of MOT on edge devices is not without its challenges. Edge devices typically possess limited computational resources, necessitating the development of highly optimized algorithms capable of delivering real-time performance under these constraints. The disparity between the computational requirements of state-of-the-art MOT algorithms and the capabilities of edge devices emphasizes a significant obstacle. To address these challenges, we propose a neural network pruning method specifically tailored to compress complex networks, such as those used in modern MOT systems. This approach optimizes MOT performance by ensuring high accuracy and efficiency within the constraints of limited edge devices, such as NVIDIA's Jetson Orin Nano. By applying our pruning method, we achieve model size reductions of up to 70% while maintaining a high level of accuracy and further improving performance on the Jetson Orin Nano, demonstrating the effectiveness of our approach for edge computing applications.

Arguably, the most crucial task of a Deep Learning based Multiple Object Tracking (MOT) is not to identify an object, but to re-identify it after occlusion. There are a plethora of trackers available to use, but not all of them have a good re-identification pipeline. In this blog post, we will focus on one such tracker, FairMOT, that revolutionised the joint optimisation of detection and re-identification tasks in tracking. The metrics that was calculated in our DeepSort post did not show good results either. The average accuracy that we got was 28.6, which is very low. FairMOT was introduced to tackle the re-identification problem.

A team of Microsoft and Huazhong University researchers this week open-sourced an AI object detector -- Fair Multi-Object Tracking (FairMOT) -- they claim outperforms state-of-the-art models on public data sets at 30 frames per second. If productized, it could benefit industries ranging from elder care to security, and perhaps be used to track the spread of illnesses like COVID-19. As the team explains, most existing methods employ multiple models to track objects: (1) a detection model that localizes objects of interest and (2) an association model that extracts features used to reidentify briefly obscured objects. By contrast, FairMOT adopts an anchor-free approach to estimate object centers on a high-resolution feature map, which allows the reidentification features to better align with the centers. A parallel branch estimates the features used to predict the objects' identities, while a "backbone" module fuses together the features to deal with objects of different scales. The researchers tested FairMOT on a training data set compiled from six public corpora for human detection and search: ETH, CityPerson, CalTech, MOT17, CUHK-SYSU, and PRW.