BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://ece.hku.hk
X-WR-CALDESC:Events for Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:20240101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250520T100000
DTEND;TZID=Asia/Hong_Kong:20250520T110000
DTSTAMP:20260509T211551
CREATED:20250603T024946Z
LAST-MODIFIED:20250603T025050Z
UID:111488-1747735200-1747738800@ece.hku.hk
SUMMARY:Keynode-Driven Dynamic Mesh Compression
DESCRIPTION:Abstract\n3D Dynamic Meshes can deliver engaging experiences in various applications\, but the storage and transmission demands associated with these data structures can be prohibitive. We address this challenge with an efficient compression technique leveraging embedded key nodes. The temporal motion of each vertex is formulated as a distance-weighted combination of transformations from neighboring key nodes\, requiring the transmission of solely the key nodes’ transformations. Through extensive experiments\, we demonstrate the effectiveness of our method in significantly reducing storage requirements while preserving the immersive quality of the visual content. \nSpeaker\nProf. Truong NGUYEN\nElectrical and Computer Engineering\,\nUniversity of California San Diego \nBiography of the Speaker\nTruong Nguyen received his B.S.\, M.S. and Ph.D. in Electrical Engineering at California Institute of Technology in 1985\, 1986 and 1989. His current research areas include 3D Human Mesh segmentation and coding as well as machine learning for medical image analysis. During his academic career\, he has published over 200 peer-reviewed journal papers\, over 380 peer-reviewed conference papers\, 1 textbook\, 3 book chapters\, and 15 issued patents. His Google H-index is 72 with over 31K citations. He received the NSF Career Award in 1995\, and IEEE Signal Processing Society’s 1992 Paper Award (student author) for the 1990 paper “Structures for M-Channel Perfect-Reconstruction FIR QMF Banks Which Yield Linear-Phase Analysis Filters” (IEEE Trans. ASSP). He received the Distinguished Teaching Award at UCSD in 2019 and three teaching awards from the ECE Dept. at UCSD in 2006\, 2008 and 2010. He is a Fellow of IEEE. \nOrganiser\nProf. S.C. CHAN\nDepartment of Electrical and Electronic Engineering\,\nThe University of Hong Kong \nCo-organiser\nIEEE Signal Processing Society Hong Kong Chapter \nAll are welcome!
URL:https://ece.hku.hk/events/20250520-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2025/06/1280-3.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20250520T140000
DTEND;TZID=Asia/Hong_Kong:20250520T150000
DTSTAMP:20260509T211551
CREATED:20250603T032657Z
LAST-MODIFIED:20250603T032657Z
UID:111558-1747749600-1747753200@ece.hku.hk
SUMMARY:Understanding Complex-Valued Transformer for Modulation Recognition
DESCRIPTION:Zoom Link: https://hku.zoom.us/j/95380440070 \nAbstract\nComplex-valued convolution neural networks (CVCNNs) have been recently applied for modulation recognition (MR)\, due to its ability to capture the relationship between the real and imaginary parts of the received signal. On the other hand\, the transformer model has been shown to be distinguished in MR by its superior capability to extract the correlation among high-dimensional signals compared to the CNN. It is a logical next step to ask whether a fully complex-valued transformer based neural network (CVTNN) can bring further performance gain? If so\, where the gain comes from? To answer these questions\, this letter designs the building blocks of the CVTNN for MR\, which is composed of a convolution embedding module\, a complete transformer encoder\, and a C2R classifier\, and establishes the estimation error bound of the proposed CVTNN from an inductive bias perspective. We theoretically prove that the estimation error bound of the proposed CVTNN is lower than that of the real-valued transformer based neural network (RVTNN) for MR. Simulation results further show that the proposed CVTNN outperforms the RVTNN and other benchmarks under different settings\, which corroborates the proposed theoretical analysis. \nSpeaker\nMr. Jingreng Lei\nDepartment of Electrical and Electronic Engineering\nThe University of Hong Kong \nSpeaker’s Biography\nJingreng Lei received the B.Eng. degree from Sun Yat-sen University\, China\, in 2023. He is currently working towards MPhil degree with The University of Hong Kong\, Hong Kong. His research interests include complex-valued neural network\, distributed optimization and wireless communication. \nAll are welcome!
URL:https://ece.hku.hk/events/20250520-3/
LOCATION:Online via Zoom
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
END:VCALENDAR