Accelerating Real-Time Video Analytics on a Heterogeneous CPU and FPGA Platform

Accelerating Real-Time Video Analytics on a Heterogeneous CPU and FPGA Platform

Databricks via YouTube Direct link

Why Real-Time ?

1 of 20

1 of 20

Why Real-Time ?

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Accelerating Real-Time Video Analytics on a Heterogeneous CPU and FPGA Platform

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Why Real-Time ?
  2. 2 Real-Time Insights
  3. 3 Existing Real-Time Analytics Solution
  4. 4 Real-Time Video Analytics
  5. 5 Main Phases of Video Analytics Pipeline
  6. 6 Architecture of CPU based Pipeline
  7. 7 Inference Phase
  8. 8 From DStreams to Structured Streaming
  9. 9 Custom Connector: reading from custom data source
  10. 10 Code Snippet
  11. 11 Deep Learning Inference
  12. 12 Performance of CPU based solution
  13. 13 Challenges with existing software-based solutions
  14. 14 Hardware Accelerators - Alternate Solution
  15. 15 Heterogenous CPU+FPGA based pipeline
  16. 16 Performance of CPU+FPGA based solution
  17. 17 Distributed System Configuration
  18. 18 Demo Application
  19. 19 Summary
  20. 20 Megh Solution Stack: reduces complexity of programming FPGA

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.