Responsible Computer Vision - Model Failures and Solutions

Responsible Computer Vision - Model Failures and Solutions

Bolei Zhou via YouTube Direct link

Benchmark Performance

4 of 24

4 of 24

Benchmark Performance

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Responsible Computer Vision - Model Failures and Solutions

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

  1. 1 Intro
  2. 2 Standard Visual Recognition Pipeline
  3. 3 Visual Recognition Benchmark
  4. 4 Benchmark Performance
  5. 5 Dataset Bias
  6. 6 Adversarial Examples
  7. 7 Benchmark Challenge Adversar
  8. 8 RobustNav Dynamics Corruptid
  9. 9 Domain Adaptation: Train on Source Test on
  10. 10 Domain Adversarial Adaptatio
  11. 11 Adapting to Imbalanced Data
  12. 12 Adaptation with Self-Training Entropy Minimization for UDA
  13. 13 SENTRY: Selective Entropy Optimization Selective Entropy Minimization
  14. 14 Selective Entropy Loss
  15. 15 SENTRY Results: Image Classification
  16. 16 SENTRY Results: MiniDomainNet
  17. 17 Extension to Semantic Segmentation
  18. 18 Performance Degradation from Bias
  19. 19 Geographic Bias
  20. 20 Does object recognition work for everyone?
  21. 21 Can domain adaptation make obj rec work for everyone?
  22. 22 Geographically diverse data
  23. 23 Additional challenges in GeoDA
  24. 24 Summary: Responsible Vision

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.