Jesse Scott
Papers
2
Total Citations
12
H-Index
2
About
Jesse Scott is a researcher at the intersection of computer vision and biomechanics, whose work bridges the gap between what we see and what we feel. His primary research focuses on inferring dynamic physical forces—specifically foot pressure and center of pressure—directly from standard video footage of human motion. This novel approach challenges traditional biomechanics, which relies on specialized force plates and pressure mats, by proposing that rich dynamic information is latent within simple kinematic data. Scott’s major contribution is the development and validation of two end-to-end deep learning architectures, PressNet and PressNet-Simple, which are designed to regress 2D foot pressure heatmaps from 2D human pose sequences. This work, detailed in his most-cited paper (7 citations), effectively learns the physics of human-ground interaction from visual appearance alone. By enabling the estimation of Center of Pressure and Base of Support from ubiquitous video, Scott’s research has significant implications for accessible gait analysis, rehabilitation monitoring, and humanoid robotics, promising to democratize biomechanical assessment beyond the lab.
Research Focus
Key Achievements
Top Papers
- 1
- 2