CONVINCE: Collaborative Cross-Camera Video Analytics at the Edge
This addresses efficiency and accuracy issues for edge-based video analytics in applications like city monitoring, though it is an incremental improvement over existing methods.
The paper tackles the problem of high computation cost, bandwidth, and privacy in dense video camera deployments by introducing CONVINCE, a collaborative cross-camera analytics system that reduces transmitted frames to ~25% while achieving ~91% object identification accuracy.
Today, video cameras are deployed in dense for monitoring physical places e.g., city, industrial, or agricultural sites. In the current systems, each camera node sends its feed to a cloud server individually. However, this approach suffers from several hurdles including higher computation cost, large bandwidth requirement for analyzing the enormous data, and privacy concerns. In dense deployment, video nodes typically demonstrate a significant spatio-temporal correlation. To overcome these obstacles in current approaches, this paper introduces CONVINCE, a new approach to look at the network cameras as a collective entity that enables collaborative video analytics pipeline among cameras. CONVINCE aims at 1) reducing the computation cost and bandwidth requirements by leveraging spatio-temporal correlations among cameras in eliminating redundant frames intelligently, and ii) improving vision algorithms' accuracy by enabling collaborative knowledge sharing among relevant cameras. Our results demonstrate that CONVINCE achieves an object identification accuracy of $\sim$91\%, by transmitting only about $\sim$25\% of all the recorded frames.