Light field denoising by sparse 5D transform domain collaborative filtering

Alain, Martin; Smolić, Aljoša

doi:10.1109/mmsp.2017.8122232

Cited by 47 publications

(37 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to assess the denoising performance of the 4D anisotropic diffusion with the proposed model on real light fields, in comparison with state of the art methods, e.g., with [2], we considered the same 12 light fields of the EPFL dataset [33]. The light fields have been extracted from the raw captures using Dansereau's Matlab light field toolbox [34] in the same conditions as in [2].…”

Section: Denoising Results With Real Light Fieldsmentioning

confidence: 99%

“…This led us to introduce a novel denoising method that we called Anisotropic Diffusion Denoising 4D (ADD4D). Denoising results show that the proposed method compares well to the best state of the art method [2], even outperforming it at high noise levels, while being considerably faster. Preliminary light field denoising results with ADD4D have been presented in [3].…”

Section: Introductionmentioning

confidence: 88%

“…In [14], the light field is denoised by stacking EPIs in a 3D volume using the VBM4D video denoising method [15], while the authors in [16] apply a hyper-fan shaped filter in a 4D discrete Fourier transform domain. The BM3D denoising algorithm [17] is extended in [2] to light fields. The only approach of regularization proposed so far for light field denoising is the one of [1] using a global variational framework integrating contributions of regularizers on vector fields computed in EPI 2D spaces.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Four-Dimensional Anisotropic Diffusion Framework With PDEs for Light Field Regularization and Inverse Problems

Allain

Guillo

Guillemot

2020

IEEE Trans. Comput. Imaging

View full text Add to dashboard Cite

In this paper, we consider the problem of vectorvalued regularization of light fields based on PDEs. We propose a regularization method operating in the 4D ray space that does not require prior estimation of disparity maps. The method performs a PDE-based anisotropic diffusion along directions defined by local structures in the 4D ray space. We analyze light field regularization in the 4D ray space using the proposed 4D anisotropic diffusion framework by first considering a light field toy example, i.e., a tesseract. This simple light field example allows an in-depth analysis of how each eigenvector influences the diffusion process. We then illustrate the diffusion effect for several light field processing applications: denoising, angular and spatial interpolation, regularization for enhancing disparity estimation as well as inpainting.

show abstract

Section: Denoising Results With Real Light Fieldsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 88%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Four-Dimensional Anisotropic Diffusion Framework With PDEs for Light Field Regularization and Inverse Problems

Allain

Guillo

Guillemot

2020

IEEE Trans. Comput. Imaging

View full text Add to dashboard Cite

show abstract

“…Five methods were chosen for comparison: the classic 2D image denoiser BM3D [13], which denoises each SAI inde- pendently; the video denoiser V-BM4D [14], which considers the LF SAIs as a video sequence, and its variation V-BM4D-EPI [15] that works on the EPI sequence; the LF denoiser HyperFan4D [18], which works on the 4D frequency space of the LF; and LFBM5D [19], which extends the BM3D model into a 5D framework. Two baseline methods were used for evaluation: Avg-All, which angularly replicates X avg to reconstruct the entire LF; APA-syn, which uses L syn from the syn-Net as denoising output.…”

Section: Resultsmentioning

confidence: 99%

Light Field Denoising via Anisotropic Parallax Analysis in a CNN Framework

Chen

Hou

Chau

2018

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

Light field (LF) cameras provide perspective information of scenes by taking directional measurements of the focusing light rays. The raw outputs are usually dark with additive camera noise, which impedes subsequent processing and applications. We propose a novel LF denoising framework based on anisotropic parallax analysis (APA). Two convolutional neural networks are jointly designed for the task: first, the structural parallax synthesis network predicts the parallax details for the entire LF based on a set of anisotropic parallax features. These novel features can efficiently capture the high frequency perspective components of a LF from noisy observations. Second, the view-dependent detail compensation network restores non-Lambertian variation to each LF view by involving view-specific spatial energies. Extensive experiments show that the proposed APA LF denoiser provides a much better denoising performance than state-of-the-art methods in terms of visual quality and in preservation of parallax details.Index Terms-Light field, anisotropic parallax feature, denoising, convolutional neural networks J. Chen, and L.-P. Chau are with the

show abstract

“…We thus propose to apply denoising as a final step. For that purpose, we use the state-of-the art LFBM5D filter introduced in [22]. This filter takes full advantage of the 4D nature of light fields by creating disparity compensated 4D patches which are then stacked together with similar 4D patches along a 5 th dimension.…”

Section: Denoisingmentioning

confidence: 99%

A Pipeline for Lenslet Light Field Quality Enhancement

Matysiak

Grogan

Pendu

et al. 2018

2018 25th IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

In recent years, light fields have become a major research topic and their applications span across the entire spectrum of classical image processing. Among the different methods used to capture a light field are the lenslet cameras, such as those developed by Lytro. While these cameras give a lot of freedom to the user, they also create light field views that suffer from a number of artefacts. As a result, it is common to ignore a significant subset of these views when doing high-level light field processing. We propose a pipeline to process light field views, first with an enhanced processing of RAW images to extract subaperture images, then a colour correction process using a recent colour transfer algorithm, and finally a denoising process using a state of the art light field denoising approach. We show that our method improves the light field quality on many levels, by reducing ghosting artefacts and noise, as well as retrieving more accurate and homogeneous colours across the sub-aperture images.

show abstract

Light field denoising by sparse 5D transform domain collaborative filtering

Cited by 47 publications

References 18 publications

Four-Dimensional Anisotropic Diffusion Framework With PDEs for Light Field Regularization and Inverse Problems

Four-Dimensional Anisotropic Diffusion Framework With PDEs for Light Field Regularization and Inverse Problems

Light Field Denoising via Anisotropic Parallax Analysis in a CNN Framework

A Pipeline for Lenslet Light Field Quality Enhancement

Contact Info

Product

Resources

About