Skip to main content
. Author manuscript; available in PMC: 2023 Apr 7.
Published in final edited form as: IEEE Trans Neural Syst Rehabil Eng. 2023 Feb 1;31:426–436. doi: 10.1109/TNSRE.2022.3226067

Fig. 2.

Fig. 2.

A. The MAGIC Table with the camera frame and webcam mounted above the table. B. Subject sits at a table with two circular targets drawn on the table surface. The subject moves the cup with the ball rolling inside from one target to the other. A web camera was mounted above the table surface to record the cup-and-ball kinematics in real time. Participants wear a white smooth glove to color-match the table surface and avoid errors in the color-based object detection. C. Cross-sectional view of the 3D-printed cup visualizes the magnet at the bottom of the cup and the metal green ball that rolls inside the orange bowl. The computer vision algorithm distinguishes objects by color to extract shapes and their position information.