Keynote: Ross Girshick - Robustness Across the Data Abundance Spectrum

Keynote: Ross Girshick - Robustness Across the Data Abundance Spectrum

Andreas Geiger via YouTube Direct link

Einleitung

1 of 25

1 of 25

Einleitung

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

Keynote: Ross Girshick - Robustness Across the Data Abundance Spectrum

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

  1. 1 Einleitung
  2. 2 Axes of Robustness
  3. 3 The Numbers You See in Papers
  4. 4 The Reality
  5. 5 Back to the Drawing Board
  6. 6 It All Starts with Naming Things
  7. 7 Naming is for Communication
  8. 8 Human Human Communication
  9. 9 Object Spotting Game car
  10. 10 Task: Detect Everything!
  11. 11 Problem 2: Data Efficient Learning
  12. 12 Problem 1: Benchmarking
  13. 13 Building LVIS
  14. 14 PASCAL VOC and COCO were "easy" to Build
  15. 15 Troubles with 1000's of Categories
  16. 16 Facing These Problems, What Can We Do?
  17. 17 Federated Dataset Design to the Rescue
  18. 18 Annotation Pipeline
  19. 19 LVIS Annotations Quality
  20. 20 Where does the vocabulary come from?
  21. 21 Data Driven Category Discovery
  22. 22 The Long Tail is Inescapable! 100
  23. 23 Results on LVIS v1 validation set
  24. 24 Best Practices - Know What's Noise
  25. 25 The LVIS Challenge at ECCV 2020!

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.