Jingying Wang
Hi, I am Jingying Wang, a fourth-year Ph.D. candidate from the Computer Science and Engineering department of the University of Michigan. I am luckily co-advised by Prof. Xu Wang from the UM CSE, and Vitaliy Popov from UM Medicine.
My research lies at the intersection of Human-Computer Interaction (HCI) and medical education, with a focus on enhancing the visual understanding in medical training. Medical procedures are inherently visual tasks that require learners to know where to look, how to interpret complex visual cues, and how to translate those perceptions into precise actions. Due to limited access to experts, trainees often learn from videos and simulations that lack personalized and interactive feedback. My work bridges this gap by designing human–AI systems that enable interactive learning experiences grounded in multimodal data sources, including video, gaze, speech, hand gestures, and etc.
[Google Scholar][CV]
News
Mar/2025
Honored to be selected to receive the Barbour Scholarship! [Article]
Sep/2024
Excited to kick off our new project, supported by the NSF award SCH: Multimodal Techniques to Enhance Intra- and Post-operative Learning and Coordination between Attending and Resident Surgeons, advised by Xu Wang, Vitaliy Popov, and Anhong Guo!
Apr/2024
Our paper “Looking Together ≠ Seeing the Same Thing: Understanding Surgeons' Visual Needs During Intra-operative Coordination and Instruction” got an honorable mention award in CHI2024
Publication
SurgGaze is an implicit calibration method that uses tool–tissue interactions as natural cues to improve surgeons’ gaze tracking. It reduces gaze error by 40.6% compared to standard calibration, enabling more reliable attention analysis in both simulated and real operating rooms. (Paper Under Review)
Jingying Wang
SurgGraph is a scene-graph pipeline for understanding laparoscopic videos by encoding expert-defined surgical relationships into LLM-generated programs grounded in segmentation and depth maps. It enables quantitative analysis of surgical processes and outperforms standard scene-graph and vision-language models in clip retrieval and question answering, providing more accurate and educationally valuable insights. (Paper Under Review)
Jingying Wang
Looking Together ≠ Seeing the Same Thing: Understanding Surgeons' Visual Needs During Intra-operative Coordination and Instruction. (CHI '24)
Xinyue Chen*, Vitaliy Popov*, Jingying Wang, Michael Kemp, Gurjit Sandhu, Taylor Kantor, Natalie Mateju, and Xu Wang
[Paper]
Real-time Facial Animation for 3D Stylized Character with Emotion Dynamics. (MM '23)
Ye Pan, Ruisi Zhang, Jingying Wang, Yu Ding, and Kenny Mitchell
[Paper]
