Papers
6
Total Citations
85
H-Index
5
About
Maxime Devanne is a leading researcher at the intersection of human motion analysis, rehabilitation robotics, and deep learning. His work focuses on developing intelligent robotic systems that can understand, assess, and guide human movement, particularly for physical rehabilitation. Devanne’s major contributions include creating multi-level motion analysis frameworks for kinaesthetic rehabilitation, where his 2017 paper (21 citations) established methods for robot coaches to evaluate exercise quality. He demonstrated real-world impact through the R-COOL randomized trial (2022, 21 citations), proving the technical feasibility of humanoid robots supervising stretching exercises for chronic low back pain—a breakthrough for home-based therapy adherence. Devanne also advanced activity recognition with hierarchical LSTM networks (2019, 17 citations) for smart home and assistive robot applications, and explored knowledge distillation in fully convolutional networks for time series classification (2022, 14 citations). His co-design approach for rehabilitation robots (2018) emphasizes error classification to improve patient-robot interaction. By generating shared latent variables for robot imitation learning (2019), Devanne bridges the gap between human movement understanding and robotic execution. His work has accumulated over 85 citations, positioning him as a key innovator in making rehabilitation more accessible through socially assistive robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6