Yongwei Li: Research project 3, “Modelling of Plenoptic capture in geometrical optics considering lightfield sampling and aliasing” (Mid Sweden University)

Project Information:

Recent years have seen a great development towards recreating visual reality from captured light field information based on different capturing models. Currently available models can be divided into two groups, i.e. wave models and ray models, each has its benefits and drawbacks. On one hand, by considering wave optics in capturing models, diffraction, PSF and other wave properties can be described at the expense of computational complexity. On the other hand, ray models which have lower model complexity suffer from a lower descriptive level of light behavior. These drawbacks make room for further research into modeling of plenoptic capture, tailored to the needs of capturing systems such as computational plenoptic imaging. In response to this, MIUN initiated work on intermediate models and approaches in plenoptic imaging systems, trying to fill the gap between wave models and ray models.

Education and Training:

Several courses, e.g. Optical Models and Digital Photography, Fundamentals of Realistic 3D and Multiple View Geometry have been completed in order to get a profound and solid background knowledge. Based on given courses, internal presentations have been given to the Realistic 3D research group at MIUN. According to training program planned by ETN-FPI, several training schools were attended. Workshop on “Parallel and Distributed Computing with Matlab” was attended December 2016, focusing on practical application of parallel computing.

Research and Development:

During the first year in Mid Sweden University, I focused on background knowledge and literature review to build up profound knowledge in stereoscopic and multi-stereoscopic systems, capture, representation and reconstruction of 3D scenes, projective geometry, computation and estimation of camera matrix, lens aberrations and 3D reconstruction of cameras and structure.

During the first secondment in Valencia, Spain, collaboration with ESR5 Gabriele Scrofani on the depth estimation for FiMic microscope has been carried out in order to tackle problems such as occlusion handling and improve the depth estimation accuracy for data captured by FiMic microscope.

Yongwei Li has a B.E. degree and an MSc Degree both in Computer Science and Technology from Liaoning Normal University, China. His research interest mainly lies in the field of computer vision, e.g., signal processing, image analysis, image retrieval and light field capturing etc.

 
Contact Information
Yongwei Li
Department of Information Systems and Technology
Mid Sweden University
Mailing Address:
Building L, Office no. L330, Holmgatan 10, 851 70, Sundsvall, Sweden
Contact Information. +46 101 428 056
Email Address: yongwei.li@miun.se