Prof. Stanley Chan named a recipient of the 2022 IEEE Signal Processing Society Best Paper Award
Stanley Chan, Elmore Associate Professor in Purdue University’s Elmore Family School of Electrical and Computer Engineering has been named a recipient of the 2022 IEEE Signal Processing Society Best Paper Award. The paper, entitled “Plug-and-Play ADMM for Image Restoration: Fixed-Point Convergence and Applications,” was published in IEEE Transactions on Computational Imaging in 2017.
Co-authors of the paper include two alumni of Purdue ECE, Dr. Xiran Wang (PhD ECE ’19) and Dr. Omar Elgendy (PhD ECE ’19). As of March 2023, the paper has been cited more than 570 times on Google Scholar.
The plug-and-play (PnP) algorithm is an important algorithmic breakthrough that offers a principled way of integrating physics models and deep learning priors. The significance of the paper is the theoretical relaxation from the previously known non-expansive criteria of deep learning prior to a boundedness criterion which is much easier to satisfy. The paper opens the door by allowing people to use any reasonably well designed deep learning denoisers as the prior, thus enabling a wide range of imaging applications in electronic imaging, photon-counting, medical imaging, and more.
Each year, the Signal Processing Society and IEEE honor a select number of members with awards for their research and work in signal processing and for their involvement in both IEEE and the Signal Processing Society. The Best Paper Award honors the author(s) of a paper of exceptional merit dealing with a subject related to the Society’s technical scope, and appearing in one of the Society’s solely owned transactions, the Journal of Selected Topics in Signal Processing, the Transactions on Computational Imaging, or the Transactions on Signal and Information Processing over Networks. Judging is based on general quality, originality, subject matter, and timeliness.