Abstract

This presentation explores the integration of deep learning and computer vision technologies in edge computing environments. The talk addresses the challenges and opportunities of deploying AI models on resource-constrained edge devices, focusing on optimization techniques that balance computational efficiency with model accuracy. Key topics include model compression methods, hardware-aware neural architecture design, and real-time inference strategies for visual perception tasks. The discussion will highlight practical applications across various domains including IoT, autonomous systems, and smart infrastructure, demonstrating how these technologies enable intelligent decision-making at the network edge without relying on cloud connectivity. Recent advances in computer vision systems will also be examined.

Bio.

Chun-Yi Lee is a Professor in the Department of Computer Science and Informa on Engineering at Na onal Taiwan University (NTU), Taipei, Taiwan, and the supervisor of Elsa Lab. He received his B.S. and M.S. degrees from Na onal Taiwan University in 2003 and 2005, respec vely, and his M.A. and Ph.D. degrees from Princeton University, Princeton, NJ, USA, in 2009 and 2013, respec vely, all in Electrical Engineering. Prof. Lee joined the Department of Computer Science at NTHU as an Assistant Professor in 2015. He was promoted to Associate Professor in 2019 and to full Professor in 2023. In 2024, Prof. Lee joined the Department of Computer Science and Informa on Engineering at Na onal Taiwan University. Prior to 2015, he served as a senior engineer at Oracle America, Inc. in Santa Clara, CA, USA, from 2012 to 2015. Prof. Lee founded Elsa Lab at Na onal Tsing Hua University in 2015. Under his leadership, Elsa Lab has garnered several pres gious awards from global robo cs and AI challenges. These include the first place at the NVIDIA Embedded Intelligent Robo cs Challenge in 2016, the first place at the NVIDIA Jetson Robo cs Challenge in 2018, the  second place in the Person-In-Context (PIC) Challenge at ECCV 2018, the second place in the NVIDIA AI at the Edge Challenge in 2020, and the Best Solu on Award (1st Place) in the Small Object Detec on Challenge for Spo ng Birds at MVA 2023.

Prof. Lee's research primarily focuses on deep reinforcement learning (DRL), intelligent robo cs, computer vision (CV), and parallel compu ng systems. His contribu ons include developing key deep learning methodologies for intelligent robo cs, such as sim-to-real and real-to-sim training, transfer techniques for robo c policies, digital twins, scene coordinate regression approaches, and domain adapta on techniques for seman c segmenta on models. He has also advanced explora on approaches for DRL agents, mulagent reinforcement learning (MARL) techniques, and genera ve modeling methodologies, including superresolu on and score-based genera ve models, as well as autonomous naviga on strategies. His work has been published at major ar ficial intelligence (AI) conferences such as NeurIPS, CVPR, IJCAI, AAMAS, ICLR, ICML, ECCV, BMVC, CoRL, ICRA, IROS, GTC, MVA, and others. Addi onally, his research appears in top AI journals, including IEEE Transac ons on Pa ern Recogni on and Machine Intelligence (TPAMI), IEEE Transac ons on Neural Networks and Learning Systems (TNNLS), Journal of Machine Learning Research (JMLR), and ACM Transac ons on Evolu onary Learning and Opmiza on (TELO). In the realm of parallel systems, Prof. Lee has introduced innova ve graphics processing unit (GPU) enhancement methodologies to improve efficiency. His work in this area has been published in IEEE Transac ons on Very Large Scale Integra on Systems (TVLSI), IEEE Transac ons on Computer-Aided Design of Integrated Circuits and Systems (TCAD), the Design Automa on Conference (DAC), the Asia and South Pacific Design Automa on Conference (ASP-DAC), and Interna onal Conference on Computer Design (ICCD). Prof. Lee's research is par cularly impac ul in autonomous systems, decision-making systems, game engines, and vision-AI based robo c applica ons. In 2024, Prof. Lee was awarded the MARC AcademiaIndustry Collabora on Excellent Research Award by MediaTek Inc., and received the Garmin Scholar Fellowship from Garmin Inc. He was honored with the Academia Sinica Early-Career Inves gator Research Achievement Award in 2022 and the Ta-You Wu Memorial Award from the Ministry of Science and Technology (MOST) in 2020. These pres gious awards in Taiwan recognize outstanding achievements in intelligence compu ng for young researchers. Prof. Lee has garnered numerous accolades, including outstanding research awards, dis nguished teaching awards, innova on teaching awards, young scholar research awards, and contribu on awards from ins tu ons such as NVIDIA Deep Learning Ins tute (DLI), The Taiwan IC Design Society (TICD), the Founda on for the Advancement of Outstanding Scholarship (FAOS), The Chinese Ins tute of Electrical Engineering (CIEE), Taiwan Semiconductor Industry Associa on (TSIA), Ins tute of Informa on & Compu ng Machinery (IICM), and Na onal Tsing Hua University (NTHU). In the academic community, Prof. Lee will serve as the Program Co-Chair at the Interna onal Conference on Machine Vision Applica ons (MVA 2025). He has served as the Area Chairs at NeurIPS 2023-2024, ICLR 2024, and ACML 2024, and as tutorial chair of MVA 2023. His roles have included commi ee member and reviewer at numerous interna onal and domes c conferences. He has been session chair and technical program commi ee member mul ple mes at IROS, ASP-DAC, NoCs, ISVLSI, and MVA, and has held various chair roles at different interna onal conferences. Prof. Lee has served as paper reviewer for NeurIPS, ICML, CVPR, ICLR, AAAI, ICCV, BMVC, ICRA, IROS, IEEE TPAMI, IEEE TVLSI, IEEE TCAD, IEEE ISSCC, and IEEE ASP-DAC on mul ple occasions. He was the main organizer of the 3rd, 4th, and 5th Augmented Intelligence and Interac on (AII) Workshops from 2019 to 2023, and chaired the ACML Workshop on Machine Learning for Mobile Robot Vision and Control (MRVC) in 2021. Prof. Lee served as the director of the NVIDIA-NTHU Joint Innova on Center from 2023 to 2024, and served as the co-director of the MOST Office for Interna onal AI Research Collabora on from 2018 to 2020. From 2024, Prof. Lee serves as the co-director of the NVIDIANTU Joint Innova on Center. Prof. Lee is a professional member of IEEE and ACM.

All faculty and students are welcome to join.