Silhouette-based 3D Human Pose Estimation Using a Single Wrist-mounted 360° Camera

In this paper, we propose a framework for 3D human pose estimation using single 360° camera mounted on the user’s wrist. Perceiving with such simple setup has remarkable potential various applications (e.g., daily-living activity monitoring, motion analysis sports training). However, no existing method tackled task due to difficulty of estimating from image in which only part body is captured, and because lack training data. We translating wrist-mounted images into poses. Since are first try task, cannot use datasets. To address issue, synthetic data build our own dataset. This solution, however, creates different problem, that domain gap between real inference. resolve silhouette-based generation created task. Extensive experiments comparing several baseline methods demonstrated effectiveness approach.