Completed
Inference invariant condition
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Learning Robust Imaging Models without Paired Data
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Outline
- 3 Linear approximation for imaging process
- 4 The error effects
- 5 Model based approaches
- 6 Deep learning (DL) based approaches
- 7 Data bottleneck in DL
- 8 Data collection in video superresolution
- 9 Goal of the talk
- 10 Image denoising
- 11 The basic idea
- 12 Model formulation
- 13 Numerical method
- 14 One remark on overfitting issue
- 15 Quantitative results for real noisy images
- 16 Qualitative results
- 17 Latent space verification
- 18 Real-world noisy images from Huawei
- 19 Image segmentation
- 20 Probabilistic model
- 21 Examples
- 22 Deep CV model
- 23 Distributions in latent space
- 24 Motivation
- 25 The case of unpaired datasets
- 26 Unpaired degradation modeling
- 27 The idea
- 28 The loss function
- 29 Inference invariant condition
- 30 Synthetic noisy images
- 31 Experiments
- 32 Visual results
- 33 Summary
