fig5

Unlocking the head: unleashing deep learning and depth camera for free head movement gaze estimation

Figure 5. Performance analysis of GPN with different variables. (A) error analysis of the horizontal coordinates of the target points; (B) Error analysis of the vertical coordinates of the target points; (C) Error analysis of the horizontal and vertical coordinates of the target points; (D) Error analysis of the spatial horizontal location of volunteers’ heads; (E) Error analysis of the spatial vertical location of volunteers’ heads; (F) Error analysis of the spatial axis location of volunteers’ heads; (G) Error analysis of the pitch angle of volunteers’ heads; (H) Error analysis of the paw angle of volunteers’ heads. The central points indicate the mean values, and the blue shaded regions represent the corresponding variance. GPN: Gaze-Point-Net.

Intelligence & Robotics
ISSN 2770-3541 (Online)

Portico

All published articles are preserved here permanently:

https://www.portico.org/publishers/oae/

Portico

All published articles are preserved here permanently:

https://www.portico.org/publishers/oae/