Light Field Super-Resolution using a Low-Rank Prior and Deep Convolutional Neural Networks

Farrugia, Reuben A.; Guillemot, Christine

doi:10.1109/tpami.2019.2893666

Cited by 38 publications

(30 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to its very deep network structure, VDSR introduce residuals to reduce the gradient vanishing problem and introduce gradient clipping to solve the gradient exploding problem. Farrugia [12] proposed an outstanding light filed SR method, by introducing deep convolutional network into light field combined with low rank, illustrating that the convolutional network can achieve better performance if combined with other classic algorithms.…”

Section: Relate Workmentioning

confidence: 99%

An Improved Feedback Network Superresolution on Camera Lens Images for Blind Superresolution

Wang

2021

Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

Most of the recent advances in image superresolution (SR) assume that the blur kernel during downsampling is predefined (e.g., Bicubic or Gaussian kernel), but it is a difficult task to make it suitable for all the realistic images. In this paper, we propose an Improved Superresolution Feedback Network (ISRFN) which is designed free to predefine the downsampling blur kernel by dealing with real-world HR-LR image pairs directly without downsampling process. We propose ISRFN by modifying the layers and network structures of the famous Superresolution Feedback Network (SRFBN). We trained the ISRFN with the Camera Lens Database named City100, which produced the HR and LR on the same lens, respectively, free for downsampling, so our proposed ISRFN is free to estimate the blur kernel. Due to different camera lens (smartphone and DSLR) databases, we perform two series of experiments under two camera lenses-based City100 databases, respectively, to choose the optimum network structures; experiments make it clear that different camera lens-based databases have different optimum network structures. We also compare our two ISRFNs with the state-of-the-art algorithms on performance; experiments show that our proposed ISRFN outperforms other state-of-the-art algorithms.

show abstract

Section: Relate Workmentioning

confidence: 99%

An Improved Feedback Network Superresolution on Camera Lens Images for Blind Superresolution

Wang

2021

Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

show abstract

“…During training, the system each time receives a 4D patch of LF, which is spatially cropped to 96 × 96 as input. For spatial SR, the downsampling is based on the classical model (Farrugia and Guillemot 2018)…”

Section: Experiments Data and Experiments Settingsmentioning

confidence: 99%

High-Order Residual Network for Light Field Super-Resolution

Meng

Liu

et al. 2020

AAAI

View full text Add to dashboard Cite

Plenoptic cameras usually sacrifice the spatial resolution of their SAIs to acquire geometry information from different viewpoints. Several methods have been proposed to mitigate such spatio-angular trade-off, but seldom make use of the structural properties of the light field (LF) data efficiently. In this paper, we propose a novel high-order residual network to learn the geometric features hierarchically from the LF for reconstruction. An important component in the proposed network is the high-order residual block (HRB), which learns the local geometric features by considering the information from all input views. After fully obtaining the local features learned from each HRB, our model extracts the representative geometric features for spatio-angular upsampling through the global residual learning. Additionally, a refinement network is followed to further enhance the spatial details by minimizing a perceptual loss. Compared with previous work, our model is tailored to the rich structure inherent in the LF, and therefore can reduce the artifacts near non-Lambertian and occlusion regions. Experimental results show that our approach enables high-quality reconstruction even in challenging regions and outperforms state-of-the-art single image or LF reconstruction methods with both quantitative measurements and visual evaluation.

show abstract

“…Another assumption for light field processing is to consider the 4D data as image sequences. Several methods are based on this assumption and apply the approaches from video processing to handle the sequences [42], [43]. However, such an assumption omits the relations between the spatial and angular coordinates.…”

Section: B High-dimensional Convolutionmentioning

confidence: 99%

LightGAN: A Deep Generative Model for Light Field Reconstruction

et al. 2020

View full text Add to dashboard Cite

A light field image captured by a plenoptic camera can be considered a sampling of light distribution within a given space. However, with the limited pixel count of the sensor, the acquisition of a high-resolution sample often comes at the expense of losing parallax information. In this work, we present a learning-based generative framework to overcome such tradeoff by directly simulating the light field distribution. An important module of our model is the high-dimensional residual block, which fully exploits the spatio-angular information. By directly learning the distribution, our approach can generate both high-quality sub-aperture images and densely-sampled light fields. Experimental results on both real-world and synthetic datasets demonstrate that the proposed method outperforms other state-of-the-art approaches and achieves visually more realistic results. INDEX TERMS Light field reconstruction, view synthesis, 4D convolution, high-dimension residual block, generative adversarial networks, deep learning, computational imaging

show abstract

Light Field Super-Resolution using a Low-Rank Prior and Deep Convolutional Neural Networks

Cited by 38 publications

References 44 publications

An Improved Feedback Network Superresolution on Camera Lens Images for Blind Superresolution

An Improved Feedback Network Superresolution on Camera Lens Images for Blind Superresolution

High-Order Residual Network for Light Field Super-Resolution

LightGAN: A Deep Generative Model for Light Field Reconstruction

Contact Info

Product

Resources

About