Focus: Querying Large Video Datasets with Low Latency and Low Cost

TL;DR

Focus is a system designed to enable low-latency, cost-effective querying of large video datasets by combining cheap indexing, specialized CNNs, and clustering to reduce processing time and resource usage.

Contribution

The paper introduces Focus, a novel system that significantly reduces GPU usage and query latency for large-scale video dataset querying by combining low-cost indexing, video-specific CNNs, and clustering techniques.

Findings

Focus uses 58X fewer GPU cycles than traditional methods.

Focus achieves 37X faster query processing compared to processing all videos.

System effectively handles large video datasets with high accuracy and low cost.

Abstract

Large volumes of videos are continuously recorded from cameras deployed for traffic control and surveillance with the goal of answering "after the fact" queries: identify video frames with objects of certain classes (cars, bags) from many days of recorded video. While advancements in convolutional neural networks (CNNs) have enabled answering such queries with high accuracy, they are too expensive and slow. We build Focus, a system for low-latency and low-cost querying on large video datasets. Focus uses cheap ingestion techniques to index the videos by the objects occurring in them. At ingest-time, it uses compression and video-specific specialization of CNNs. Focus handles the lower accuracy of the cheap CNNs by judiciously leveraging expensive CNNs at query-time. To reduce query time latency, we cluster similar objects and hence avoid redundant processing. Using experiments on video streams from traffic, surveillance and news channels, we see that Focus uses 58X fewer GPU cycles than running expensive ingest processors and is 37X faster than processing all the video at query time.