University of Hawaii

Electrical Engineering

Group News

20-08, UH News featured an article about my state-of-the-art iterative AI system (recently published online in IEEE Trans. Pattern Anal. Mach. Intell., the most prestigious AI and computer vision journal (IF '20: 19.420)): AI innovation may improve COVID-19 diagnosis.
20-07, My full paper on fast and convergent iterative neural network is published online in IEEE Trans. Pattern Anal. Mach. Intell., the top journal in AI and computer vision (IF '20: 19.420). The developed framework, Momentum-Net, can rapidly solve a wide range of inverse problems in imaging, image processing, and computer vision, and achieve good generalization even when trained from limited samples. My group and collaborators are tailoring Momentum-Net for diverse low-level processing and high-level vision applications in medical imaging and computational photography. @Momentum-Net team members, thank you for all your hard work!
20-07, Emerging Technology Ventures (ETV) and the Chun's imaging and data science group created an industry-academia partnership. I look forward to solving big challenges together and developing cutting-edge "autonomous systems"/AI technologies with ETV!
20-06, I am hosting a talk entitled "Robust Robotic Assembly under Uncertainty" at UHM EE. Anirban Sinha, a Ph.D. candidate of ME at SUNY Stony Brook, will give the talk @1pm on 6/24. Please email me if you didn't receive the email announcement that gives a Zoom link or want to chat with him on 6/24 (after 2pm HST).
20-05, UH CoE wrote an article on my graduate course EE616: Computational Image Processing and Computer Vision of which final project, dental image diagnostics, is partnered with Hawaiʻi Dental Service: Engineering students develop real-world solutions through industry-integrated courses. See also UH News. I am proud that many students successfully completed EE616 coursework in S2020 that has unprecedented challenges due to COVID-19!
20-05, Full paper on improved low-count PET reconstruction with an iterative neural network (collab. w/ Dr. Hongki Lim (Siemens)) is published online in IEEE Trans. Med. Imag., the top medical imaging journal (IF '20: 9.710). Congrats on paper acceptance and graduation, Hongki! Best wishes for your next adventure in Siemens!
20-05, Invited paper on joint light-field image reconstruction and depth estimation (collab. w/ Zhengyu Huang (UM)) is presented at IEEE Intl. Conf. on Acoust., Speech, and Signal Process. (ICASSP) as an invited lecture.
20-02, My group has an opportunity immediately available for a Postdoctoral Fellow in autonomous driving. See here for the position overview, qualification, and application instruction. [20-05,06: We are screening/interviewing candidates. 20-07: The position is filled.]
20-02, UH News wrote a news article on our recent light-field photography work using transparent photodetectors and advanced data science solution (published in Nature Photonics (2-yr. IF: 31.583)): Engineering professor’s 3D camera improves research in Hawaiʻi.

19-12, Invited paper on incorporating hand-crafted filters in convolutional autoencoders (collab. w/ Caroline Crockett (UM) and Dr. David Hong (UPenn)) is presented at IEEE Intl. Workshop on Compt. Adv. in Multi-Sensor Adaptive Process. (CAMSAP).
19-09, My full paper on fast and convergent convolutional analysis operator learning (CAOL) is published online in IEEE Trans. Image Process., the top journal in image processing and ML/AI (IF '20: 8.980). The paper proposes a fast and convergent multi-block optimization framework and introduces the link between CAOL and training convolutional neural networks.