Based on Epipolar Plane Images, a Maximum Likelihood Approach for Depth Field Estimation PROJECT TITLE : A Maximum Likelihood Approach for Depth Field Estimation Based on Epipolar Plane Images ABSTRACT: From dense picture arrays, this work presents a multi-resolution approach for determining depth. Hand-held plenoptic cameras have been made possible by recent advancements in consumer electronics. These systems use a micro-lens array placed in front of the imaging sensor on the focal point of the primary camera lens to capture numerous perspectives of a scene in a single shot. These views can be combined to produce more accurate depth maps when processed together. The computational complexity of global optimization strategies based on match cost functions is reduced by making a local estimate based on the maximising of total log-likelihood spatial density aggregated along the epipolar lines corresponding to each view pair, in this contribution. This technique uses epipolar plane pictures to estimate the depth field using the local maximum likelihood method. A multi-resolution approach is employed to address the ambiguity problem that emerges on flat surface regions while maintaining bandwidth in the vicinity of edges. In actuality, the depth map resolution is diminished in areas where maximising the higher resolution functional is not well-trained.. The key advantages of the proposed system are its reduced computational complexity and its excellent accuracy in estimating depth. There is a reasonable balance between accuracy, robustness, and discontinuity handling in the suggested scheme. Did you like this research project? To get this research project Guidelines, Training and Code... Click Here facebook twitter google+ linkedin stumble pinterest Based on CNN Feature Learning, a Local Metric for Defocus Blur Detection Using Conventional Monomodal Normal Atlases, a New Multi-Atlas Registration Framework for Multimodal Pathological Images