Learning Light Field Angular Super-Resolution via a Geometry-Aware Network

The acquisition of light field images with high angular resolution is costly. Although many methods have been proposed to improve the angular resolution of a sparsely-sampled light field, they always focus on the light field with a small baseline, which is captured by a consumer light field camera. By making full use of the intrinsic \textit{geometry} information of light fields, in this paper we propose an end-to-end learning-based approach aiming at angularly super-resolving a sparsely-sampled light field with a large baseline. Our model consists of two learnable modules and a physically-based module. Specifically, it includes a depth estimation module for explicitly modeling the scene geometry, a physically-based warping for novel views synthesis, and a light field blending module specifically designed for light field reconstruction. Moreover, we introduce a novel loss function to promote the preservation of the light field parallax structure. Experimental results over various light field datasets including large baseline light field images demonstrate the significant superiority of our method when compared with state-of-the-art ones, i.e., our method improves the PSNR of the second best method up to 2 dB in average, while saves the execution time 48$\times$. In addition, our method preserves the light field parallax structure better.

[1]  Ravi Ramamoorthi,et al.  Local light field fusion , 2019, ACM Trans. Graph..

[2]  Qionghai Dai,et al.  Light Field Reconstruction Using Deep Convolutional Network on EPI , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  John Flynn,et al.  Stereo magnification , 2018, ACM Trans. Graph..

[4]  Pierre Vandergheynst,et al.  Tensor low-rank and sparse light field photography , 2016, Comput. Vis. Image Underst..

[5]  Ting-Chun Wang,et al.  Learning-based view synthesis for light field cameras , 2016, ACM Trans. Graph..

[6]  Sven Wanner,et al.  Datasets and Benchmarks for Densely Sampled 4D Light Fields , 2013, VMV.

[7]  Gordon Wetzstein,et al.  Compressive light field photography using overcomplete dictionaries and optimized projections , 2013, ACM Trans. Graph..

[8]  Marc Levoy,et al.  High performance imaging using large camera arrays , 2005, ACM Trans. Graph..

[9]  Christine Guillemot,et al.  A Framework for Learning Depth From a Flexible Subset of Dense and Sparse Light Field Views , 2019, IEEE Transactions on Image Processing.

[10]  Zhibo Chen,et al.  Light Field Spatial Super-Resolution Using Deep Efficient Spatial-Angular Separable Convolution , 2019, IEEE Transactions on Image Processing.

[11]  Bastian Goldlücke,et al.  A Dataset and Evaluation Methodology for Depth Estimation on 4D Light Fields , 2016, ACCV.

[12]  Aggelos K. Katsaggelos,et al.  Compressive Light Field Sensing , 2012, IEEE Transactions on Image Processing.

[13]  Jie Chen,et al.  Light Field Denoising via Anisotropic Parallax Analysis in a CNN Framework , 2018, IEEE Signal Processing Letters.

[14]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[15]  Xiaoming Chen,et al.  Fast Light Field Reconstruction with Deep Coarse-to-Fine Modeling of Spatial-Angular Clues , 2018, ECCV.

[16]  Frédo Durand,et al.  Light Field Reconstruction Using Sparsity in the Continuous Fourier Domain , 2014, ACM Trans. Graph..

[17]  W. Freeman,et al.  Understanding Camera Trade-Offs through a Bayesian Analysis of Light Field Projections , 2008, ECCV.

[18]  Ashok Veeraraghavan,et al.  Light field denoising, light field superresolution and stereo camera based refocussing using a GMM light field patch prior , 2012, 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[19]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[20]  Pavan K. Turaga,et al.  Compressive Light Field Reconstructions Using Deep Learning , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[21]  In-So Kweon,et al.  Accurate depth map estimation from a lenslet light field camera , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Tieniu Tan,et al.  End-to-End View Synthesis for Light Field Imaging with Pseudo 4DCNN , 2018, ECCV.

[23]  Robert Bregovic,et al.  Light Field Reconstruction Using Shearlet Transform , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[24]  Qionghai Dai,et al.  Learning Sheared EPI Structure for Light Field Reconstruction , 2019, IEEE Transactions on Image Processing.

[25]  In-So Kweon,et al.  Learning a Deep Convolutional Network for Light-Field Image Super-Resolution , 2015, 2015 IEEE International Conference on Computer Vision Workshop (ICCVW).

[26]  Yi Yang,et al.  Occlusion Aware Unsupervised Learning of Optical Flow , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[27]  Feng Liu,et al.  Video Frame Interpolation via Adaptive Separable Convolution , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[28]  Xiaoou Tang,et al.  Learning a Deep Convolutional Network for Image Super-Resolution , 2014, ECCV.