Search for: All records

Real-time video analytics typically require video frames to be processed by a query to identify objects or activities of interest while adhering to an end-to-end frame processing latency constraint. This imposes a continuous and heavy load on backend compute and network infrastructure. Video data has inherent redundancy and does not always contain an object of interest for a given query. We leverage this property of video streams to propose a lightweight Load Shedder that can be deployed on edge servers or on inexpensive edge devices co-located with cameras. The proposed Load Shedder uses pixel-level color-based features to calculate a utility score for each ingress video frame and a minimum utility threshold to select interesting frames to send for query processing. Dropping unnecessary frames enables the video analytics query in the backend to meet the end-to-end latency constraint with fewer compute and network resources. To guarantee a bounded end-to-end latency at runtime, we introduce a control loop that monitors the backend load and dynamically adjusts the utility threshold. Performance evaluations show that the proposed Load Shedder selects a large portion of frames containing each object of interest while meeting the end-to-end frame processing latency constraint. Furthermore, it does not impose a significant latency overhead when running on edge devices with modest compute resources. 

Over the last 20 years, mobile computing has evolved to encompass a wide array of increasingly data-rich applications. Many of these applications were enabled by the Cloud computing revolution, which commoditized server hardware to support vast numbers of mobile users from a few large, centralized data centers. Today, mobile's next stage of evolution is spurred by interest in emerging technologies such as Augmented and Virtual Reality (AR/VR), the Internet of Things (IoT), and Autonomous Vehicles. New applications relying on these technologies often require very low latency response times, increased bandwidth consumption, and the need to preserve privacy. Meeting all of these requirements from the Cloud alone is challenging for several reasons. First, the amount of data generated by devices can quickly saturate the bandwidth of backhaul links to the Cloud. Second, achieving low-latency responses for making decisions on sensed data becomes increasingly difficult the further mobile devices are from centralized Cloud data centers. And finally, regulatory or privacy restrictions on the data generated by devices may require that such data be kept locally. For these reasons, enabling next-generation technologies requires us to reconsider the current trend of serving applications from the Cloud alone.