US 11,734,855 B2
Rotation equivariant orientation estimation for omnidirectional localization
Chao Zhang, Cambridge (GB); Ignas Budvytis, Cambridge (GB); and Stephan Liwicki, Cambridge (GB)
Assigned to Kabushiki Kaisha Toshiba, Tokyo (JP)
Filed by Kabushiki Kaisha Toshiba, Tokyo (JP)
Filed on Nov. 20, 2020, as Appl. No. 17/100,036.
Prior Publication US 2022/0164986 A1, May 26, 2022
Int. Cl. G06K 9/00 (2022.01); G06T 7/77 (2017.01)
CPC G06T 7/77 (2017.01) [G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer implemented method for pose estimation of an image sensor, the method comprising:
receiving an omnidirectional image of a scene captured by an image sensor;
using a trained neural network to generate a rotation equivariant feature map from said omnidirectional image of the scene; and
determining information relating to the pose of said image sensor when capturing the scene from said rotation equivariant feature map,
wherein the rotation equivariant feature map is a SO(3) indexed feature map.