High-dimensional Dense Residual Convolutional Neural Network for Light Field Reconstruction

Abstract

We consider the problem of high-dimensional light field reconstruction and develop a learning-based framework for spatial and angular super-resolution. Many current approaches either require disparity clues or restore the spatial and angular details separately. Such methods have difficulties with non-Lambertian surfaces or occlusions. In contrast, we formulate light field super-resolution (LFSR) as tensor restoration and develop a learning framework based on a two-stage restoration with 4-dimensional (4D) convolution. This allows our model to learn the features capturing the geometry information encoded in multiple adjacent views. Such geometric features vary near the occlusion regions and indicate the foreground object border. To train a feasible network, we propose a novel normalization operation based on a group of views in the feature maps, design a stage-wise loss function, and develop the multi-range training strategy to further improve the performance. Evaluations are conducted on a number of light field datasets including real-world scenes, synthetic data, and microscope light fields. The proposed method achieves superior performance and less execution time comparing with other state-of-the-art schemes.

arXiv comment14 pages. IEEE Transactions on Pattern Analysis and Machine Intelligence (2019)

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• TensorFlow: A system for large-scale machine learning
• Learning-Based View Synthesis for Light Field Cameras
• Light Field Reconstruction Using Convolutional Network on EPI and Extended Applications
• Light Field Super-Resolution Via Graph-Based Regularization
• Spatial and Angular Resolution Enhancement of Light Fields Using Convolutional Neural Networks
• Partial light field tomographic reconstruction from a fixed-camera focal stack

Cited by in corpus (10)

• Disentangling Light Fields for Super-Resolution and Disparity Estimation
• Light Field Image Super-Resolution with Transformers
• Spatial-Angular Attention Network for Light Field Reconstruction
• LRT: An Efficient Low-Light Restoration Transformer for Dark Light Field Images
• Light Field View Synthesis via Aperture Disparity and Warping Confidence Map
• Unsupervised Light Field Depth Estimation via Multi-view Feature Matching with Occlusion Prediction
• Spatial-Angular Interaction for Light Field Image Super-Resolution
• High-Order Residual Network for Light Field Super-Resolution
• Light Field Implicit Representation for Flexible Resolution Reconstruction
• Efficient Light Field Reconstruction via Spatio-Angular Dense Network