Dr. Jiang Junjun won the Outstanding Doctoral Dissertation Award of China Computer Federation
After intense competition, Jiang Junjun's paper "Face Super-resolution via Consistent Manifold Learning" won the Outstanding Doctoral Dissertation Award of China Computer Federation in 2016, and achieved the goal of Excellent Doctoral Thesis for CCF. This is another breakthrough for Computer School, marking the quality of college graduate training has entered the domestic advanced ranks.
This paper focuses on the three major challenges faced by the learning-based super-resolution method for face detection: the high-dimensional space formed by finite high-dimensional data samples does not have local linearity, and the manifold space is susceptible to noise and high-low resolution images Space manifold geometry structure is inconsistent, the face of the library to carry out the expression theory, the human face image visual representation theory, high and low resolution manifold keep learning theory. Which solves a number of key problems in learning super-resolution technology, and provides a new solution to the problem of super-resolution reconstruction of human face in video surveillance.
Dr. Jiang Junjun in 2009 to enter the NERCMS for a master's degree, the same year through the "1+4" applications, in 2011 to enter the PhD program. Dr. Jiang Junjun focuses on super-resolution face detection technology in surveillance video.
Under the guidance of instructor Prof. Hu Ruimin, Dr. Jiang Junjun has published 25 papers in IEEE Transactions on Image Processing, IEEE Transactions on Multimedia, IEEE Transactions on Circuits and the International Conference on Computer Science and Technology, IEEE Transactions on Video Technology Circuits and Systems, IEEE Communications and ACM International Multimedia Conference, International Conference on Artificial Intelligence, and American Artificial Intelligence Conference, and won Best Student Paper Award in International Multimedia Modeling Conference. The paper has been cited 429 times by Google Academic including the proposed method of local constraint (95 cited) and multi-layer iterative neighbor embedding (74 cited) , which have been recognized and followed by domestic and foreign counterparts .
It is understood that the Outstanding Doctoral Dissertation Award of China Computer Federation is produced by the special award committee, and the annual winners no more than 10, which is representing the the highest level of doctoral education.