English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent advances in neural networks (NNs) have enabled automatic querying oflarge volumes of video data with high accuracy. While these deep NNs canproduce accurate annotations of an object's position and type in video, theyare computationally expensive and require complex, imperative deployment codeto answer queries. Prior work uses approximate filtering to reduce the cost ofvideo analytics, but does not handle two important classes of queries,aggregation and limit queries; moreover, these approaches still require complexcode to deploy. To address the computational and usability challenges ofquerying video at scale, we introduce BlazeIt, a system that optimizes queriesof spatiotemporal information of objects in video. BlazeIt accepts queries viaFrameQL, a declarative extension of SQL for video analytics that enablesvideo-specific query optimization. We introduce two new query optimizationtechniques in BlazeIt that are not supported by prior work. First, we developmethods of using NNs as control variates to quickly answer approximateaggregation queries with error bounds. Second, we present a novel searchalgorithm for cardinality-limited video queries. Through these theseoptimizations, BlazeIt can deliver up to 83x speedups over the recentliterature on video processing.

BlazeIt: Optimizing Declarative Aggregation and Limit Queries for Neural Network-Based Video Analytics