Unmasking the Inductive Biases of Unsupervised Object Representations for Video Sequences

Unmasking the Inductive Biases of Unsupervised Object Representations for Video Sequences

Andreas Geiger via YouTube Direct link

Intro

1 of 25

1 of 25

Intro

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

Unmasking the Inductive Biases of Unsupervised Object Representations for Video Sequences

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

  1. 1 Intro
  2. 2 Collaborators
  3. 3 Why Object-Centric Learning? Explicit object representation
  4. 4 Tracking by Detection
  5. 5 Unsupervised Object-Centric Learning
  6. 6 Common Principle
  7. 7 Categorization of Approaches
  8. 8 VIMON: Attention Network
  9. 9 VIMON: Next-Frame Prediction
  10. 10 TBA Tracker Array
  11. 11 TBA Mid-Level Representation
  12. 12 TBA Spatial Transformation
  13. 13 Spatial Mixture Models: OP3
  14. 14 OP3 Dynamics Network
  15. 15 CLEAR MOT Metrics
  16. 16 Datasets
  17. 17 Results on SpMOT
  18. 18 How Well Do Models Accumulate Evidence Over Time?
  19. 19 Dependency of Performance on Number of Objects
  20. 20 Challenging Cases
  21. 21 VMDS Challenge Sets
  22. 22 Out-of-Distribution Test Sets
  23. 23 Runtime Analysis Runtime on Single RTX 2080 TI GPU
  24. 24 Conclusions
  25. 25 2D Annotations

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.