Back to results
US 12,762,731 B2Granted
Methods for refining RGBD camera poses
AssigneeMagic Leap, Inc.
InventorXiaolin Wei
Life of this patent
Estimating the term…
Abstract
A method for refining poses includes receiving a plurality of poses and computing a relative pose set by determining a first set of relative poses between image frame pairs for a first subset of the image frame pairs having a temporal separation between image frames of the image frame pairs less than a threshold, and determining a second set of relative poses between image frame pairs for a second subset of the image frame pairs having a temporal separation between image frames of the image frame pairs greater than the threshold.
Claims20
A method for refining poses, the method comprising:
- capturing a plurality of image frames using a depth camera;
- computing a relative pose set by:
- determining a first set of relative poses between image frame pairs for a first subset of the image frame pairs having a temporal separation between image frames of the image frame pairs less than a threshold;
- and
- determining a second set of relative poses between image frame pairs for a second subset of the image frame pairs having a temporal separation between image frames of the image frame pairs greater than the threshold,
- wherein the first subset of image frame pairs and the second subset of image frame pairs together comprise every image frame pair for the plurality of image frames;
- performing a first relative pose optimization process for the first set of relative poses;
- and
- performing a second relative pose optimization process different from the first relative pose optimization process for the second set of relative poses.
6 more elements
A method for refining poses, the method comprising:
- capturing a plurality of depth image frames using a depth camera;
- computing a plurality of poses, each of the plurality of poses being associated with one of the plurality of depth image frames;
- computing a first portion of a relative pose set by performing a first pose optimization process for a first set of relative poses between depth image frame pairs having a temporal separation between depth image frames of the depth image frame pairs less than a threshold;
- and
- computing a second portion of the relative pose set by performing a second pose optimization process for a second set of relative poses between the depth image frame pairs having a temporal separation between the depth image frames of the depth image frame pairs greater than the threshold, wherein the second pose optimization process comprises a different initialization than the first pose optimization process, and wherein the depth image frame pairs having temporal separation less than the threshold and the depth image frame pairs having temporal separation greater than the threshold together comprise every depth image frame pair for the plurality of depth image frames;
- performing volumetric reconstruction for a portion of the relative pose set to produce a global geometry;
- and
- refining each pose of the plurality of poses with respect to the global geometry to produce a plurality of refined poses.
5 more elements
Dependencies are inferred from the claim text.