Abstract: Thanks to the advances of deep learning in recent years, computer vision has found new applications in other disciplines, such as biomedical applications, material identification, and IC fabrication. Though important, several of such vision-based cross-disciplinary research topics are still in their early stage and were rarely explored due to the difficulties and complexities in data collection, problem formulation, etc. This talk will summarize our experiences in applying data-driven computer vision to IC-Fab EDA and THz imaging.
First, fabricating IC wafers involves a multiple-step sequence of photolithographic and chemical etch processing, leading to serious inconsistencies between the designed layout patterns and their corresponding circuit shapes, thereby causing IC defects (e.g., broken wires, bridges, and enclosures). This has brought up an important topic of IC-Fab EDA: Design for Manufacturability (DfM). This talk will first show how computer vision can effectively and efficiently address the following three essential issues for IC DfM: (1) How to predict the manufactured IC circuit shape from an IC layout in a pre-simulation process and (2) How to optimize photomask patterns so that fabricated IC circuit shape can well match their design patterns.
Second, THz computational imaging has recently attracted significant attention thanks to its non-invasive, non-destructive, non-ionizing, material-classification, and ultra-fast nature for object exploration and inspection. However, its strong water absorption nature and low noise tolerance lead to undesired blurs and distortions of reconstructed THz images. The performances of existing methods are highly constrained by the diffraction-limited THz signals.
In this talk, we will show how to break the limitations of THz imaging with the aid of rich spectral amplitude and phase information carried in prominent THz frequencies for THz image restoration and reconstruction. To this end, we propose a novel physics-guided DNN model that fuses multi-spectral features of THz images for high-performance THz computed tomography.
About the Speaker: Prof. Chia-Wen Lin is currently a Professor at the Department of Electrical Engineering, National Tsing Hua University (NTHU), Taiwan. He also serves as Deputy Director of the AI Research Center of NTHU. He was a Visiting Professor at the Graduate School of Informatics, Kyoto University, Japan in 2023. His research interests include image/video processing, computer vision, and video networking.
Dr. Lin is an IEEE Fellow and served on IEEE Circuits and Systems Society (CASS) Fellow Evaluating Committee during 2021 till 2023. He also serves as IEEE CASS BoG Members-at-Large (2022~2024). He was the Steering Committee Chair of IEEE ICME (2020-2021), IEEE CASS Distinguished Lecturer (2018~2019), and President of the Chinese Image Processing and Pattern Recognition (IPPR) Association, Taiwan (2019~2020). He has served as Associate Editor of IEEE Transactions on Image Processing, IEEE Transactions on Multimedia, IEEE Transactions on Circuits and Systems for Video Technology, and IEEE Multimedia. He also served as a Steering Committee member of the IEEE Transactions on Multimedia. He was Chair of the Multimedia Systems and Applications Technical Committee of the IEEE CASS. He served as TPC Chair of IEEE ICME 2010, IEEE ICIP 2019, and PCS 2022, and the Conference Chair of IEEE VCIP in 2018. His papers won the Best Paper Award of IEEE VCIP 2015, and the Young Investigator Award of VCIP 2005.
Venue: EEE Executive Seminar Room (S2.2-B2-53)