Dr. Jie Chen, Assistant Professor in the Department of Computer Science, whose research interests include computational photography, computer vision, and novel imaging systems and their image processing systems (ISP) pipelines, has aspirations in working out vision problems with tools of mathematics and machine learning.

First obtaining his bachelor’s and master’s degrees from the School of Optical and Electronic Information at Huazhong University of Science and Technology, Dr Chen then moved to Singapore. There, he completed his PhD studies at the Nanyang Technological University (NTU), following which he became a post-doctoral researcher at the ST Engineering-NTU Corporate Laboratory, and then a Senior Algorithm Engineer at Omnivision Technologies Inc., Singapore. In 2019, Dr Chen joined HKBU through the University’s “Talent 100” initiative.

Visual signals, a focus of Dr Chen’s research and an important part of computer vision, are high-dimensional spaces involving several dimensions. High-dimensional spaces include the spectral and time dimensions. Dr Chen’s aim is to use an omnidimensional framework to develop a visual engine to model all dimensions.

In addition, Dr Chen is attempting to use deep learning methods to reconstruct light fields for applications such as film editing, 3D avatar capture, 3D objects, and 3D environmental mapping. Humans can only see a limited part of the hyperspectral domain, so Dr Chen is using an RGB camera to reconstruct hyperspectral images, allowing people to look freely from within, and even beyond the limit of camera based-lines.

Using Novel Methods for Universal Application

A keen lover of mathematics and physics since high school, Dr Chen is passionate about using novel methods and Artificial Intelligence to help bring about new potentials within traditional software. Dr Chen wishes to utilise multiple novel methods in his research in order to produce widely-available equipment with intelligent algorithms for anyone to create high-class imagery.

At HKBU, Dr Chen is incredibly pleased to be conducting his research surrounded by friendly colleagues. “In my experience, the relationship between academic researchers can be very fraught due to intense competition,” Dr Chen said. “However, the colleagues I have met at HKBU have been extremely helpful, and the same goes for the senior University staff - I am very glad to have so much support from all around the University.”

Source: Research Office

× Visual comparisons of different methods on 2x reconstruction. The figures labelled “Ours” are the visual reconstruction generated by Dr Chen’s team. (Source: Henry Yeung, Junhui Hou, Xiaoming Chen, Jie Chen, Zhibo Chen, and Yuk Ying Chung, “Light Field Spatial Super-Resolution Using Deep Spatial-Angular Interleaved CNN,” under major revision in IEEE Transactions on Image Processing (IEEE TIP), vol. 28, no. 5, pp. 2319-2330, 2019)