2. The Paper
The Perception-Distortion Tradeoff, CVPR 2018 Oral Presentation
Yochai Blau and Tomer Michaeli, Technion – Israel Institue of Technology
Conclusion
The Perception-Distortion Tradeoff Exists
Algorithms Cannot Achieve both Low Distortion & Good Perceptual Quality
We need to pay more attention to Evaluate Image Processing Results
11. Definition: Distortion
▷ How Similar to Ground Truth?
▷ Expectation over the Joint Distribution 𝑃 𝑋, 𝑋
▷ Full Reference Metric
𝔼 Δ 𝑋, 𝑋
• Mean Squared Error
• SSIM
Image Quality Assessment: From Error Visibility to Structural Similarity, TIP 2004
• MS-SSIM
Multiscale Structure Similarity for Image Quality Assessment, CSSC 2004
• IFC
An Information Fidelity Criterion for Image Quality Assessment using Natural Scene Statistics, TIP 2005
• VIF
Image Information and Visual Quality, TIP 2006
• Perceptual Loss
Perceptual Losses for Real-time Style Transfer and Super Resolution, ECCV 2016
13. Definition: Perceptual Quality
The Degree to Which it Looks Like a
Natural Image
𝑑𝑖𝑣(𝑝 𝑋, 𝑝 𝑋)
Perceptual Quality
∝ Human Mean Opinion Score
∝ No-Reference Metric
∝ Real & Fake Test
∝ Divergence in GANs
14. Definition: Perceptual Quality
▷ The Degree to Which it Looks Like a Natural Image
▷ Perceptual Quality
∝ Human Mean Opinion Score
∝ No-Reference Metric
∝ Real & Fake Test
∝ Divergence in GANs
𝑑𝑖𝑣(𝑝 𝑋, 𝑝 𝑋)
• Total Variation
• Jenson-Shannon Divergence
Generative Adversarial Nets, NIPS 2014
• Wasserstein Distance
Wasserstein GAN, ICML 2017
• Any f-Divergence
f-GAN, NIPS 2016
15. Definition: Perceptual Quality
▷ The Degree to Which it Looks Like a Natural Image
▷ Perceptual Quality
∝ Human Mean Opinion Score
∝ No-Reference Metric (In Experiments)
∝ Real & Fake Test
∝ Divergence in GANs
𝑑𝑖𝑣(𝑝 𝑋, 𝑝 𝑋)
• BRISQUE
No-Reference Image Quality Assessment in the Spatial Domain, TIP 2012
• BLIINDS-II
Blind Image Quality Assessment: A Natural Scene Stastics Approach in the DCT Domain, TIP 2012
• NIQE
Making a Completely Blind Image Quality Analyzer, IEEE SP Letters, 2013
• Ma et al.
Learning a No-Reference Quality Metric for Single Image Super-Resolution, CVIU 2017
22. Definition 1: The Perception-Distortion Function
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
23. Definition 1: The Perception-Distortion Function
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
DistortionPerception
Recon.
Algorithm
24. Definition 1: The Perception-Distortion Function
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
DistortionPerception
Recon.
Algorithm
25. Definition 1: The Perception-Distortion Function
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
DistortionPerception
Recon.
Algorithm
26. Definition 1: The Perception-Distortion Function
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
DistortionPerception
Recon.
Algorithm
27. Definition 1: The Perception-Distortion Function
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
DistortionPerception
Recon.
Algorithm
28. Theorem 1: The Perception-Distortion Tradeoff
If 𝑑(𝑝 𝑋, 𝑝 𝑋) is Convex in 𝑝 𝑋,
𝑃 𝐷 is Monotonically Non-Increasing & Convex
29. Theorem 1: The Perception-Distortion Tradeoff
If 𝑑(𝑝 𝑋, 𝑝 𝑋) is Convex in 𝑝 𝑋,
𝑃 𝐷 is Monotonically Non-Increasing & Convex
Small Distortion+ Large Perceptual Quality Loss
30. Theorem 1: The Perception-Distortion Tradeoff
If 𝑑(𝑝 𝑋, 𝑝 𝑋) is Convex in 𝑝 𝑋,
𝑃 𝐷 is Monotonically Non-Increasing & Convex
Small Perceptual Quality Gain Large Distortion ++
31. Theorem 1: The Perception-Distortion Tradeoff: Proof
If 𝑑(𝑝 𝑋, 𝑝𝑋) is Convex in 𝑝𝑋
𝑃 𝐷 is Monotonically Non-Increasing & Convex
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
32. Theorem 1: The Perception-Distortion Tradeoff: Proof
If 𝑑(𝑝 𝑋, 𝑝𝑋) is Convex in 𝑝𝑋
𝑃 𝐷 is Monotonically Non-Increasing & Convex
𝑃 𝐷 = min
𝑝𝑋|𝑌
𝑑(𝑝 𝑋, 𝑝𝑋) 𝑠. 𝑡. 𝔼 Δ 𝑋, 𝑋 ≤ 𝐷
𝑃 𝐷 = Minimum 𝑑(𝑝 𝑋, 𝑝𝑋) ∈ 𝐷 𝐷
If 𝐷 Increases, 𝐷 𝐷 Increases, 𝑃(𝐷) is Non-Increasing
33. Theorem 1: The Perception-Distortion Tradeoff: Proof
If 𝑑(𝑝 𝑋, 𝑝𝑋) is Convex in 𝑝𝑋
𝑃 𝐷 is Monotonically Non-Increasing & Convex
𝜆𝑃 𝐷1 + 1 − 𝜆 𝑃 𝐷2 ≥ 𝑃(𝜆𝐷1 + 1 − 𝜆 𝐷2)
53. • Rate Distortion Function 𝑅(𝐷)
• If D is a tolerable distortion,
then 𝑅 𝐷 is the Minimum Rate with which the data source can be coded
• Rate ∝ Bit per Second ∝ Source Quality
Slide from Y. Blau
55. • The Optimal Algorithm is Application Dependent
• eg ) Medical Image / Personal Photos
• One Cannot Dominate the Others
Slide from http://www.screenplaysmag.com/2013/11/25/keys-to-solving-the-strategic-challenge-posed-by-msos-reliance-on-mpeg-2/
60. Theorem 2
You will Never need to Degrade
More than 3dB in PSNR (or x2 in MSE)
to obtain Perpect Perceptual Quality
Slide from Y. Blau
61. If Condition is Satisfied, PD Tradeoff arises in Other Domain
𝐷𝑖𝑠𝑡𝑜𝑟𝑡𝑖𝑜𝑛 = 𝔼 Δ 𝑋, 𝑋
𝑃𝑒𝑟𝑐𝑒𝑝𝑡𝑖𝑜𝑛 = 𝑑𝑖𝑣(𝑝 𝑋, 𝑝 𝑋)
Theorem 1: The Perception-Distortion Tradeoff
If 𝑑(𝑝 𝑋, 𝑝 𝑋) is Convex in 𝑝 𝑋,
𝑃 𝐷 is Monotonically Non-Increasing & Convex