#ieee-signal-processing-magazine #construction #communications #artificial-intelligence #processing #networking

 

Modern computational imaging systems capture rich visual information across diverse modalities and scales, yet translating raw measurements into accurate, interpretable representations remains a fundamental challenge. This talk presents our recent advances in physics-driven machine learning for computational imaging, spanning three interconnected themes: reconstruction, analysis, and generation. In the reconstruction thread, we explore novel view synthesis, 3D scene understanding, and high-fidelity sparse-view reconstruction. In the analysis thread, we discuss how to leverage physical prior and how foundation models can be grounded with physical constraints for robust generalisation. In the generation thread, we present physics-aware approaches for reflection removal and shadow removal that disentangle physically distinct image formation components. Across all themes, a unifying philosophy emerges: embedding domain knowledge of underlying optical physics into the learning framework yields models that are more accurate, data-efficient, and interpretable.

 

 

Dr. Bihan Wen is currently an Associate Professor, Provost's Chair in Artificial Intelligence for Computational Imaging, Director of the CISS-ROSE Research Centre at Nanyang Technological University, Singapore. He received the MS and PhD degrees in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign (UIUC), USA, in 2015 and 2018, respectively. His research interests include artificial intelligence, computational imaging, image and video processing. Dr. Wen is an IEEE Senior Member. He currently serves as a member of the IEEE Visual Signal Processing and Communications TC (2022-present) and IEEE Image, Video, and Multidimensional Signal Processing TC (2026-present), and as Vice Chair of the IEEE Signal Processing Society Singapore Chapter. He is currently an Associate Editor for IEEE Signal Processing Magazine (2026-Present), Guest Editor for IEEE TMM and Pattern Recognition. He was a member (2018-2021) and SubCommittee Chair (2021-2024) of the IEEE Computational Imaging TC, served as Associate Editor for IEEE T-CSVT (2020-2026), and as Conference Program Chair for IEEE MIPR 2022, IEEE ICARCV 2024, IEEE VCIP 2026, and IEEE ICARCV 2026. He regularly serves as Area Chair for IEEE ICIP, ICASSP, ICME, ISCAS, and AAAI. Dr. Wen received the 2026 Graduate Student Mentoring Award, 2025 Research Young Faculty Award (Runner-Up), and 2022 Early-Career Teaching Excellence Award from NTU. He received the 2026 Advanced Imaging Early Career Award from CIOP, and 2023 IEEE VSPC Rising Star (Runner-Up) from IEEE VSCP. He received the Best Paper Awards at IEEE ICIEA 2023 and IEEE MIPR 2023, and Best Paper Award (Runner-Up) from IEEE ICME 2020. He has been ranked among the World Top 2% Scientists in Artificial Intelligence by Stanford University and the Top 0.05% Highly Cited Scholar by ScholarGPS.