|
Instructor:
Prof.
Gene
Cheung
Lectures: T/R
11:30-13:00
Location:
(see
info in eClass)
Announcement
- 1/01/2026: eClass page is online.
- 12/23/2025: Class homepage is online.
Course
Summary
Fundamental image processing theories and
algorithms. Signal representations using transforms
and wavelets are reviewed. Signal reconstruction
methods using total variation, sparse coding and
low-rank prior, based on convex optimization, are
discussed. Applications include image compression,
restoration, and enhancement. Prior background in
digital signal processing (EECS 4452 or equivalent)
is required; basics in linear algebra and convex
optimization are strongly recommended!
Required
Textbook
- M. Vetterli, J. Kovacevic, V. Goyal, Foundations
of Signal Processing, Cambridge
University Press, 2014. (also available online HERE)
Supplementary
Material
- R. Gonzalez, R. Woods, Digital
Image Processing (4th Edition), Pearson
Education Limited, 2018.
- S. Boyd, L. Vandenberghe, Introduction
to Applied Linear Algebra: Vectors, Matrices,
and Least Squares, Cambridge
University Press, 2018.
- M. Elad, Sparse
and
Redundant Representations, Springer,
2010.
- A. Ortega,
Introduction
to Graph Signal Processing, Cambridge
University Press, 2022. (available in Amazon HERE)
- G. Cheung,
E. Magli, Graph
Spectral Image Processing,
Wiley-ISTE, 2021. (available in Amazon HERE)
Key
References
- A. Beck, M. Teboulle,
"Fast
Gradient-Based Algorithms for Constrained
Total Variation Image Denoising and
Deblurring," IEEE
Transactions on Image Processing,
vol. 18, no. 11, November 2009, pp.
2419-2434.
-
M.
Aharon,
M. Elad, A. Bruckstein, "K-SVD:
An
Algorithm for designing overcomplete
sparse representation," IEEE
Transactions on Signal Processing,
vol. 54, no. 11, Nov. 2006, pp. 4311-4322.
-
E.
Candes,
X. Li, Y. Ma, J. Wright, "Robust
Principal Component Analysis?"
vol. 58, no. 3, article 11, Journal of
the ACM, May 2011.
-
S.
Boyd
et al., "Distributed Optimization and
Statistical Learning via the Alternating
Direction Method of Multipliers," Foundation
and Trends in Machine Learning, vol.
3, no. 1, January 2011, pp.1-122.
-
N.
Parikh
and S. Boyd, "Proximal Algorithms,"
Foundations and Trends in Optimization,
vol. 1, no. 3, 2013, pp. 127-239.
- J.
Han,
A. Saxena, V. Melkote, K. Rose, "Jointly
Optimized
Spatial Prediction and Block Transform for
Video and Image Coding," IEEE
Transactions on Image Processing,
vol.21, no.4, April 2012, pp. 1874--1884.
- A.
Ortega
et al., "Graph
Signal Processing: Overview, Challenges,
and Applications," Proceedings
of the IEEE, vol. 106, no.5, May
2018, pp. 808-828.
- G.
Cheung
et al., "Graph
Spectral Image Processing," Proceedings of
the IEEE, vol. 106, no. 5, May
2018, pp. 907-930.
- V.
Monga, Y. Li and Y. C. Eldar, "Algorithm
Unrolling: Interpretable, Efficient Deep
Learning for Signal and Image Processing,"
in IEEE Signal Processing Magazine,
vol.38, no.2, pp.18-44, March 2021.
-
N.
Juniusevic, A. Khalilian-Gourtani and Y.
Wang, "CDLNet: Noise-Adaptive
Convolutional Dictionary Learning
Network for Blind Denoising and
Demosaicing," in IEEE Open
Journal of Signal Processing, vol.3,
pp.196-211, 2022.
-
Y.
Yu et al., “White-box transformers via
sparse rate reduction,” the
Thirty-Seventh Annual Conference on
Neural Information Processing Systems
(NeurIPS), December 2023.
-
T.
T. Do, P. Eftekhar, S. A. Hosseini, G.
Cheung, P. Chou, "Interpretable
Lightweight Transformer via Unrolling of
Learned Graph Smoothness Priors," the
Thirty-Eighth Annual Conference on Neural
Information Processing Systems (NeurIPS),
Vancouver, Canada, December 2024.
Evaluation
- Bi-weekly assignments (40%)
- Midterm (30%)
- Course project (30%)
Course
Outline
(subject to change)
- Week 1: Linear Algebra Review
- Week 2: Inner-product, Hilbert Space
- Week 3: Image Analysis: Transforms
- Week 4: Image Analysis: Wavelets
- Week 5: Sparse Signal Representations I
- Week 6: Sparse Signal Representations II
- Week 7: Reading Week
- Week 8: Low-Rank Representations
- Week 9: Image Compression
- Week 10: Image Denoising
- Week 11: Graph Signal Processing
- Week 12: Deep Algorithm Unrolling
- Week 13: Graph Algorithm Unrolling
|
