Lingyu Du

I am a Ph.D. candidate in the Embedded Systems Group of the Computer Science Departments at Delft University of Technology, the Netherlands, supervised by Dr. Guohao Lan and Prof. Dr. Koen Langendoen. I obtained my Master’s and Bachelor’s degrees (Honors Program) in Control Science and Engineering from Harbin Institute of Technology, Harbin, China, in 2021 and 2019, respectively.

My research lies in the intersection of pervasive systems and machine learning, with a focus on enhancing efficiency and trustworthiness of machine learning systems and developing human-centered applications.

Office: 2.W.560, Building 28, Van Mourik Broekmanweg 6, Delft
Email: Lingyu.Du [AT] tudelft.nl

I am currently in the job market.

Email  /  Scholar  / LinkedIn  /   Github

News

• [2025/8/29] I am invited to be on the TPC of DATE 2026 (Track A6: Applications of AI Systems).
• [2025/7/9] One paper is accepted to appear in ACM IMWUT/UbiComp 2025.
• [2025/5/15] I am invited to be on the program committee of UbiComp/ISWC 2025 (Posters & Demos).
• [2025/5/7] I presented our paper at CPS-IOT Week 2025 in Irvine, USA.
• [2025/5/6] Our paper EfficientGaze is accepted by ACM TOSN.
• [2025/2/25] Our paper SecureGaze has been accepted to appear in ACM SenSys 2025.
• [2024/10] I presented our work at UbiComp 2024 and had a very nice spring vacation in Australia!
• [2024/8/5] Our paper PrivateGaze has been accepted to appear in ACM IMWUT/UbiComp 2024.

Publications

A resource-efficient gaze estimation framework that significantly reduces the reliance on labeled data and improve the computational efficiency for inference and calibration.

The first work that investigates the backdoor vulnerability of gaze estimation models and proposes a framework to identify backdoored gaze estimation models.

An approach to transform raw images to obfuscated images that do not contain private attributes of users but can be used for effective gaze estimation by black-box models.

An approach that achieve comparable performance with the supervised baseline, while enabling up to 6.81 and 1.67 times speedup in calibration and inference, respectively.

A eye-tracking dataset in VR, combining high-frame-rate periocular videos and high-frequency gaze data to enable accurate, multimodal emotion recognition.

A distributed active power and reactive power coordination scheme for access point voltage support.

Professional Services

• Design, Automation and Test in Europe Conference (DATE), Technical Programme Committee, 2026
  
• UbiComp / ISWC (Posters & Demos), Program Committee, 2025
  
• UbiComp / ISWC (Briefs & Notes), Reviewer, 2025
  
• Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), Reviewer, 2024
  
• ACM Transactions on Sensor Networks (TOSN), Reviewer, 2024
  
• IEEE Transactions on Human-Machine Systems, Reviewer, 2024, 2025
  
• IEEE International Conference on Mobile Ad-Hoc and Smart Systems (MASS), Reviewer, 2022, 2023, 2024
  
• Workshop on Smart Wearable Systems and Applications (conjunction with MobiCom), Publicity Chair, 2022, 2025