BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.16.0//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:20220101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;VALUE=DATE:20231220
DTEND;VALUE=DATE:20231221
DTSTAMP:20260513T073657
CREATED:20231206T081937Z
LAST-MODIFIED:20250114T075922Z
UID:17870-1703030400-1703116799@ece.hku.hk
SUMMARY:RPG Seminar – Direct Data-Driven Control Methods: From Linear Systems to Nonlinear Systems
DESCRIPTION:Data-driven control (DDC) is a control strategy that develops controllers from data for unknown systems\, and it is increasingly popular in various research fields. DDC can be divided into indirect and direct approaches. The former uses data to identify system models and subsequently designs controllers based on those models\, while the latter simplifies the process by designing controllers directly from data\, skipping the system identification step. In this seminar\, we will explore different types of direct DDC methods for linear and nonlinear systems. Initially\, we will focus on linear systems\, particularly those with unmeasurable states. In this situation\, an output feedback controller will be designed solely using input-output data. Next\, the discussion will shift towards linear systems impacted by disturbances\, where a data-driven H-infinity controller will be developed. Furthermore\, we will extend these methods to nonlinear systems\, with a critical aspect involving the introduction of the piecewise affine system as a connecting element. \nZoom Link :\nhttps://hku.zoom.us/j/96557319827?pwd=TUlSRnA0ZFI3UC9Fc0Z2VlZzWEV1Zz09\nMeeting ID: 965 5731 9827\nPassword: 477912 \nBiography of the speaker:\n\nMr. Kaijian Hu received his B.Eng. degree in automation from Liaoning University of Science and Technology and his M.Eng. in control theory and control engineering from Dalian University of Technology. Currently\, he is pursuing his Ph.D. in the Department of Electrical and Electronic Engineering at the University of Hong Kong. His primary research interests are data-driven control and unmanned aerial vehicle (UAV) control. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-direct-data-driven-control-methods-from-linear-systems-to-nonlinear-systems/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20231220T103000
DTEND;TZID=Asia/Hong_Kong:20231220T113000
DTSTAMP:20260513T073657
CREATED:20231213T063716Z
LAST-MODIFIED:20250114T075611Z
UID:17877-1703068200-1703071800@ece.hku.hk
SUMMARY:RPG Seminar – Dynamic Sparse Dataflow Architecture for Event-based Vision Inference
DESCRIPTION:Event-based vision represents a paradigm shift in how vision information is captured and processed. By only responding to dynamic intensity changes in the scene\, event-based sensing produces far less data than conventional frame-based cameras\, promising to springboard a new generation of high-speed\, low-power machines for edge intelligence. However\, processing such dynamically sparse input originated from event cameras efficiently in real time\, particularly with complex deep neural networks (DNN)\, remains a formidable challenge. Existing solutions that employ GPUs and other frame-based DNN accelerators often struggle to efficiently process the dynamically sparse event data\, missing the opportunities to improve processing efficiency with sparse data. To address this\, we propose ESDA\, a composable dynamic sparse dataflow architecture that allows customized DNN accelerators to be constructed rapidly on FPGAs for event-based vision tasks. ESDA is a modular system that is composed of a set of parametrizable modules for each network layer type. These modules share a uniform sparse token-feature interface and can be connected easily to compose an all-on-chip dataflow accelerator on FPGA for each network model. ESDA achieves substantial speedup and improvement in energy efficiency across different applications\, and it allows much wider design space for real-world deployments. \nZoom Link :\nhttps://hku.zoom.us/j/96114813885 \nBiography of the speaker:\n\nYizhao Gao received his B.Eng. degree at the University of Chinese Academy of Sciences\, he is pursuing Ph.D. degree at the University of Hong Kong. His research interests focus on reconfigurable computing and event-based vision processing. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-dynamic-sparse-dataflow-architecture-for-event-based-vision-inference/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20231220T150000
DTEND;TZID=Asia/Hong_Kong:20231220T160000
DTSTAMP:20260513T073657
CREATED:20231213T084541Z
LAST-MODIFIED:20250114T075349Z
UID:17880-1703084400-1703088000@ece.hku.hk
SUMMARY:RPG Seminar – Parallel imaging reconstruction using spatial nulling maps
DESCRIPTION:Parallel imaging is widely used in clinical MRI to accelerate data acquisition or correct artifacts through the use of multiple receiving coils where each coil exhibits a unique spatial coil sensitivity map. Parallel reconstruction using null operations (PRUNO) is a k-space reconstruction method where a k-space nulling system is derived using null-subspace bases of the calibration matrix. ESPIRiT reconstruction extends the PRUNO subspace concept by exploiting the linear relationship between signal-subspace bases and spatial coil sensitivity characteristics\, yielding a hybrid-domain approach. Yet it requires empirical eigenvalue thresholding to mask the coil sensitivity information and is sensitive to signal- and null-subspace division. In this study\, we combine the concepts of null-subspace PRUNO and hybrid-domain ESPIRiT to provide a more robust reconstruction method that extracts null-subspace bases of calibration matrix to calculate image-domain spatial nulling maps (SNMs). The proposed reconstruction method eliminates the need for coil sensitivity masking and is relatively insensitive to subspace separation\, presenting a robust parallel imaging reconstruction procedure in practice. \nZoom Link :\nhttps://hku.zoom.us/j/91980394611\nMeeting ID: 919 8039 4611 \nBiography of the speaker:\n\nJiahao Hu received his B.E. degree from the Southern University of Science and Technology in 2020. He is currently pursuing a Ph.D. in EEE department at the University of Hong Kong. His research interests include analytical reconstruction algorithms\, data-driven and model-based deep learning methods for improving biomedical imaging quality and efficiency. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-parallel-imaging-reconstruction-using-spatial-nulling-maps/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20231220T160000
DTEND;TZID=Asia/Hong_Kong:20231220T170000
DTSTAMP:20260513T073657
CREATED:20231213T084213Z
LAST-MODIFIED:20250114T075441Z
UID:17879-1703088000-1703091600@ece.hku.hk
SUMMARY:RPG Seminar – Pushing the limits of ultra-low-field MRI by dual-acquisition super-resolution
DESCRIPTION:Recent development of ultra-low-field (ULF) MRI presents opportunities for low-power\, EMI shielding-free\, and portable clinical applications of MRI. However\, the imaging performance of these emerging ULF MRI scanners remains limited due to its three orders of magnitude weaker main magnetic field\, resulting in the poor signal-to-noise ratio. Advancements in deep learning have opened new frontiers for improving ULF MRI image quality. In this seminar\, we present a novel dual-acquisition deep learning method for enhancing spatial resolution and suppressing noise/artifacts of 3D ULF brain MRI images acquired at our custom-built 0.055T brain MRI scanner.. \nZoom Link :\nhttps://hku.zoom.us/j/98172884162?pwd=VWVmeE5DWjhvcGJIeUZGdTBtRWdzQT09 \nBiography of the speaker:\n\nMan Hin Lau (Vick) obtained his MEng degree in Biomedical Engineering from Imperial College London in 2019. After a year working as a research assistant at HKU\, he is now pursuing a PhD degree with Prof Ed X Wu in the Department of Electrical and Electronic Engineering. His research focuses on the application of deep learning techniques to MRI image processing and reconstruction. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-pushing-the-limits-of-ultra-low-field-mri-by-dual-acquisition-super-resolution/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20231220T170000
DTEND;TZID=Asia/Hong_Kong:20231220T180000
DTSTAMP:20260513T073657
CREATED:20231213T084857Z
LAST-MODIFIED:20250114T075316Z
UID:17881-1703091600-1703095200@ece.hku.hk
SUMMARY:RPG Seminar – Deep learning enabled fast 3D brain MRI at 0.055 tesla
DESCRIPTION:In recent years\, there has been an intensive development of portable ultralow-field magnetic resonance imaging (MRI) for low-cost\, shielding-free\, and point-of-care applications. However\, its quality is poor and scan time is long. We propose a fast acquisition and deep learning reconstruction framework to accelerate brain MRI at 0.055 tesla. The acquisition consists of a single average three-dimensional (3D) encoding with 2D partial Fourier sampling\, reducing the scan time of T1- and T2-weighted imaging protocols to 2.5 and 3.2 minutes\, respectively. The 3D deep learning leverages the homogeneous brain anatomy available in high-field human brain data to enhance image quality\, reduce artifacts and noise\, and improve spatial resolution to synthetic 1.5-mm isotropic resolution. Our method overcomes low-signal barrier\, reconstructing fine anatomical structures that are reproducible within subjects and consistent across two protocols. It enables fast and quality whole-brain MRI at 0.055 tesla\, with potential for widespread biomedical applications. \nZoom Link :\nhttps://hku.zoom.us/j/93224346406\nMeeting ID: 932 2434 6406 \nBiography of the speaker:\n\nChristopher Man received his bachelor degree in the University of Hong Kong and is currently pursuing PhD in the University of Hong Kong\, under the supervision of Prof. Ed X. Wu. His research interests include MRI image reconstruction and deep learning. \nAll are welcome.
URL:https://ece.hku.hk/events/rpg-seminar-deep-learning-enabled-fast-3d-brain-mri-at-0-055-tesla/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
END:VCALENDAR