Home
I am Kaixu Chen (陈凯旭), a researcher at the Center for Computational Sciences (CCS), University of Tsukuba, Japan. I received my Ph.D. in Engineering from the University of Tsukuba in March 2025, where I worked with Prof. Yoshihiro Kuroda at the Life Engineering Lab.
My research lies at the intersection of computer vision, deep learning, and human motion analysis. I develop video-based methods for quantitative movement assessment, with applications in clinical gait analysis, sports analytics, and multi-view 3D human pose estimation. A recurring theme is making visual AI more robust, interpretable, and practical outside controlled laboratory environments.
Research Interests
- Clinical gait analysis and medical AI — video-based assessment of adult spinal deformity, phase-aware motion modeling, and clinically informed attention.
- 3D human motion and sports analytics — calibration-free fusion of multiple pose streams from moving cameras, with an emphasis on outdoor skiing.
- Driver behavior and human-centered vision — measurement of head-motion strategies under constrained visibility.
- Multisensory interaction — collaborative work on AI-driven visual-to-thermal feedback.
Learn more on the Research page.
Selected Recent Publications
- Chen et al., Clinical knowledge-guided attention for gait-based adult spinal deformity assessment from monocular video. Journal of Medical Imaging, 13(5), 054002, 2026.
- Wang, Chen, Ienaga, and Kuroda, Incision trajectory tracing for electrosurgical navigation by CNN-based knife contacting frames extraction method. Computer Methods and Programs in Biomedicine, 286, 109612, 2026.
- Chen et al., CanonFuse3D: Human-Centric Canonical Fusion for Calibration-Free Dual-View 3D Pose Estimation in Skiing Videos. MMSports 2026 (ACM Multimedia Workshop), accepted for oral presentation.
- Chen et al., PhaseMix: A Periodic Motion Fusion Method for Adult Spinal Deformity Classification. IEEE Access, 2024.
See the full Publications page or my Google Scholar profile.
News
- Oct 2026 — Our clinical knowledge-guided gait analysis study was published in Journal of Medical Imaging. Article
- Aug 2026 — Our electrosurgical navigation study was published online in Computer Methods and Programs in Biomedicine. Article
- Aug 2026 — CanonFuse3D was accepted at MMSports 2026, co-located with ACM Multimedia 2026, with an oral presentation scheduled. Paper
- Apr 2026 — Our clinical knowledge-guided attention framework appeared in SPIE Medical Imaging 2026: Image Processing. Proceedings paper
- Jan 2026 — FilterNet appeared in the ICEIC 2026 proceedings. Paper
- Dec 2025 — Co-authored Feel What You See, demonstrated at SIGGRAPH Asia 2025 Emerging Technologies. Paper
Last updated: October 2026.