Soshi Shimada
Papers
1
Total Citations
21
H-Index
1
About
Soshi Shimada is a leading researcher in computer vision and 3D human motion capture, with a focus on reconstructing realistic human poses and interactions from monocular video. His most cited work, "HULC: 3D Human Motion Capture with Pose Manifold Sampling and Dense Contact Guidance" (2022, 21 citations), introduces a novel framework that combines pose manifold sampling with dense contact cues to address the ambiguity of single-view motion capture. This approach enables more physically plausible and temporally coherent human animations, particularly in challenging scenarios involving occlusions or complex ground contacts. Shimada’s contributions are pivotal for applications in augmented reality, robotics, and virtual character animation, where accurate and robust human pose estimation is essential. By integrating geometric constraints with learned priors, his work bridges the gap between data-driven methods and physical realism. With a growing citation impact and a reputation for advancing the state of the art in human motion understanding, Shimada is recognized as an emerging voice in the field, pushing the boundaries of how machines perceive and reconstruct human movement.
Research Focus
Key Achievements
Top Papers
- 1