Real-Time Analytics on Computer Vision Data (DeveloperWeek Global 2020)

Discover how Rockset's indexing database reduces data latency and enables real-time analytics on high-volume computer vision data, making it possible to make decisions based on production outcomes.

Key takeaways
  • Data challenges in computer vision include high volume, velocity, and reproducibility, requiring fast processing and query latency.
  • Rockset is an indexing database that reduces data latency by not requiring ETLs or cleaning, and allows for querying of raw data as soon as it is produced.
  • Rockset’s ingest rate is very high, capable of processing gigabytes of data per second.
  • Query lambdas in Rockset allow for serverless development of applications and reduce the need for complex SQL logic.
  • Rockset supports various data formats, including JSON, and allows for building and refining knowledge graphs.
  • Real-time analytics on computer vision data requires fast querying and processing of high-volume data to make decisions based on production outcomes.
  • Standard Cognition uses Rockset to process computer vision data and build applications such as autonomous checkout.
  • The company aims to reduce data latency and increase the speed of their applications by using Rockset.