Method and computer readable medium for pose estimation
AssigneeSeiko Epson Corporation
InventorsAkshaya MISHRA, Rouzbeh MAANI
Life of this patent
Estimating the term…
Abstract
A method for pose estimation includes receiving an image containing an object, a first pose of the object in the image, and 3D boundary features of a model corresponding to the object; computing a first pose confidence of the first pose based on the image, the 3D boundary features, and the first pose; perturbing the first pose to obtain a second pose; computing a second pose confidence of the second pose based on the image, the 3D boundary features, and the second pose; determining if the second pose confidence is greater than the first pose confidence, and; and outputting the second pose and the second pose confidence if the second pose confidence is greater than the first pose confidence.
Claims16
A method that is performed by one or more processors, comprising:
- (a) receiving an image containing an object, a first pose of the object in the image, and 3D boundary features of a model corresponding to the object;
- (b) computing a first pose confidence of the first pose based on the image, the 3D boundary features, and the first pose;
- (c) perturbing the first pose to obtain a second pose;
- (d) computing a second pose confidence of the second pose based on the image, the 3D boundary features, and the second pose;
- (e) determining if the second pose confidence is greater than the first pose confidence;
- and
- (f) outputting the second pose and the second pose confidence if the second pose confidence is greater than the first pose confidence.
A non-transitory computer readable medium that embodies instructions that cause one or more processors to perform object detection, the instructions causing the one or more processors to:
- (a) receive an image containing an object, a first pose of the object in the image, and 3D boundary features of a model corresponding to the object;
- (b) compute a first pose confidence of the first pose based on the image, the 3D boundary features, and the first pose;
- (c) perturb the first pose to obtain a second pose;
- (d) compute a second pose confidence of the second pose based on the image, the 3D boundary features, and the second pose;
- (e) determine if the second pose confidence is greater than the first pose confidence;
- and
- (f) output the second pose and the second pose confidence if the second pose confidence is greater than the first pose confidence.
Dependencies are inferred from the claim text.