Geometry-Consistent Light Field Super-Resolution via Graph-Based Regularization - 2018 PROJECT TITLE :Geometry-Consistent Light Field Super-Resolution via Graph-Based Regularization - 2018ABSTRACT:Light field cameras capture the 3D information in a very scene with one exposure. This special feature makes lightweight field cameras very appealing for a selection of applications: from post-capture refocus to depth estimation and image-based rendering. But, light-weight field cameras suffer by style from strong limitations in their spatial resolution. Off-the-shelf super-resolution algorithms aren't ideal for light field information, as they do not take into account its structure. On the other hand, the few super-resolution algorithms explicitly tailored for light-weight field knowledge exhibit significant limitations, like the need to hold out a expensive disparity estimation procedure with sub-pixel precision. We have a tendency to propose a new light-weight field super-resolution algorithm meant to deal with these limitations. We tend to use the complementary data in the various lightweight field views to reinforce the spatial resolution of the full lightweight field without delay. In particular, we have a tendency to show that coupling the multi-view approach with a graph-based mostly regularizer, that enforces the sunshine field geometric structure, permits to avoid the necessity of a particular and pricey disparity estimation step. Intensive experiments show that the new algorithm compares favorably to the state-of-the-art methods for lightweight field super-resolution, both in terms of visual quality and in terms of reconstruction error. Did you like this research project? To get this research project Guidelines, Training and Code... Click Here facebook twitter google+ linkedin stumble pinterest Generalization of the Dark Channel Prior for Single Image Restoration - 2018 Her2Net A Deep Framework for Semantic Segmentation and Classification of Cell Membranes and Nuclei in Breast Cancer Evaluation - 2018
