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

To efficiently process visual data at scale, researchers have proposed twotechniques for lowering the computational overhead associated with theunderlying deep learning models. The first approach consists of leveraging aspecialized, lightweight model to directly answer the query. The secondapproach focuses on filtering irrelevant frames using a lightweight model andprocessing the filtered frames using a heavyweight model. These techniquessuffer from two limitations. With the first approach, the specialized model isunable to provide accurate results for hard-to-detect events. With the secondapproach, the system is unable to accelerate queries focusing on frequentlyoccurring events as the filter is unable to eliminate a significant fraction offrames in the video.  In this paper, we present THIA, a video analytics system for tackling theselimitations. The design of THIA is centered around three techniques. First,instead of using a cascade of models, it uses a single object detection modelwith multiple exit points for short-circuiting the inference. This earlyinference technique allows it to support a range of throughput-accuracytradeoffs. Second, it adopts a fine-grained approach to planning and processesdifferent chunks of the video using different exit points to meet the user'srequirements. Lastly, it uses a lightweight technique for directly estimatingthe exit point for a chunk to lower the optimization time. We empirically showthat these techniques enable THIA to outperform two state-of-the-art videoanalytics systems by up to 6.5X, while providing accurate results even onqueries focusing on hard-to-detect events.

THIA: Accelerating Video Analytics using Early Inference and Fine-Grained Query Planning