Image deblurring and denoising using color priors

By Neel Joshi, C. Lawrence Zitnick, Richard Szeliski, David J. Kriegman, Microsoft Research. 2009

233 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Ming-Hsuan Yang
2. Luc Van Gool
3. Lei Zhang
4. Dacheng Tao
5. Wangmeng Zuo
6. Chunhua Shen
7. Xiaogang Wang
8. Chen Change Loy

In Media Technology

1. Wangmeng Zuo
2. Ming-Hsuan Yang
3. Lei Zhang
4. Li Xu
5. Dilip Krishnan
6. Jiaya Jia
7. Sunghyun Cho
8. Jue Wang

Citations in the corpus

• Fast Image Deconvolution using Hyper-Laplacian Priors (2009, methodology): “[7] have applied them to non-blind deconvolution problems u sing IRLS to solve for the deblurred image.”
• Two-Phase Kernel Estimation for Robust Motion Deblurring (2010, background): “[20] incorporated a local two-color prior to suppress noise.”
• A Holistic Approach to Cross-Channel Image Noise Modeling and Its Application to Image Denoising (2016, unknown): “In spite of insufficient information to estimate image noise, their method showed good performance when applied to noise removal [19] and image deblurring [13].”
• Stripformer: Strip Transformer for Fast Image Deblurring (2022, unknown): “Some works utilize prior assumptions to remove nonuniform blur [3,4,5].”
• A Generalized Iterated Shrinkage Algorithm for Non-convex Sparse Coding (2013, methodology): “Several typical algorithms include iteratively reweighted least squares (IRLS) [12, 14, 23, 24, 28], iteratively reweighted 1 -minimization (IRL1) [8], iteratively thresholding method (ITM-(cid:2) p ) [33, 34], and look-up table (LUT) [25].”
• A no-reference metric for evaluating the quality of motion deblurring (2013, methodology): “• Two-color priors [Joshi et al. 2009] assume that, in natural images, colors within a local image neighborhood are linear combinations of a primary and secondary color.”
• Single Image Deblurring Using Motion Density Functions (2010, methodology): “In most recent work, image blur is modeled as the convolution of an unobserved latent image with a single, spatially invariant blur kernel [1,2,3,4,5,6,7,8,9,10,11].”