Methods, systems, and computer readable media for visual odometry using rigid structures identified by antipodal transform
View Patent ↗The subject matter described herein includes methods for visual odometry using rigid structures identified by an antipodal transform. One exemplary method includes receiving a sequence of images captured by a camera. The method further includes identifying rigid structures in the images using an antipodal transform. The method further includes identifying correspondence between rigid structures in different image frames. The method further includes estimating motion of the camera based on motion of corresponding rigid structures among the different image frames.
1. A method for hybrid motion model computation from matching feature points, the method comprising;
identify feature points in a current image frame and at least two previous image frames captured by a camera;
for a triplet of image frames including the current image frame and a previous two image frames, estimate inter-frame camera position at each frame based on matching of the feature points across the frames in the triplet; and
for the triplet of image frames, using line correspondences in the frames to compute a line-based trifocal tensor and an estimation of inter-frame camera position at every frame, wherein the line-based trifocal tensor is based on a line whose inter-frame rotation and translation are tracked and used to determine inter-frame rotation and translation of the camera for the triplet of image frames.