Abstract
The extensive rise of high-definition CCTV camera footage has stimulated both the data compression and the data analysis research fields. The increased awareness of citizens to the vulnerability of their private information, creates a third challenge for the video surveillance community that also has to encompass privacy protection. In this paper, we aim to tackle those needs by proposing a deep learning-based object tracking solution via compressed domain residual frames. The goal is to be able to provide a public and privacy-friendly image representation for data analysis. In this work, we explore a scenario where the tracking is achieved directly on a restricted part of the information extracted from the compressed domain. We utilize exclusively the residual frames already generated by the video compression codec to train and test our network. This very compact representation also acts as an information filter, which limits the amount of private information leakage in a video stream. We manage to show that using residual frames for deep learning-based object tracking can be just as effective as using classical decoded frames. More precisely, the use of residual frames is particularly beneficial in simple video surveillance scenarios with non-overlapping and continuous traffic.
Highlights
According to Cisco’s Visual Networking Index report in 2017, global data consumption has been increasing exponentially for the past decade, with video data accounting for 80% of the worldwide traffic1
The two experiments were tested on four detector/tracker combinations for both image representations: YOLOv4/Kalman-IOU tracker (KIOU), YOLOv4/ Simple Online and Realtime Tracking (SORT), tiny YOLOv4/KIOU and tiny YOLOv4/SORT
For the Higher Order Tracking Accuracy (HOTA) metric, we obtained an average score of 35.92% when residual frames were the input versus an average score of 41.87% when decoded frames were the input
Summary
According to Cisco’s Visual Networking Index report in 2017, global data consumption has been increasing exponentially for the past decade, with video data accounting for 80% of the worldwide traffic. One of the largest growing types of video data consumption is video surveillance traffic, which is set to achieve a seven-fold increase by 2022 to account for a total of 3% of the worldwide Internet traffic. This substantial surge in video surveillance data had created three major needs. The first major need is to be able to transfer and store the data which calls for the use of innovative video compression codecs. The second major need is to be able to analyze the large flow of data, which calls for the use of machine learning and deep learning algorithms.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
