Egomotion using assorted features

摘要

A system and method are disclosed for estimating camera motion of a visual input scene using points and lines detected in the visual input scene. The system includes a camera server comprising a stereo pair of calibrated cameras, a feature processing module, a trifocal motion estimation module and an optional adjustment module. The stereo pair of the calibrated cameras and its corresponding stereo pair of camera after camera motion form a first and a second trifocal tensor. The feature processing module is configured to detect points and lines in the visual input data comprising a plurality of image frames. The feature processing module is further configured to find point correspondence between detected points and line correspondence between detected lines in different views. The trifocal motion estimation module is configured to estimate the camera motion using the detected points and lines associated with the first and the second trifocal tensor.

主权项

1. A computer implemented method for estimating camera motion of a visual input scene, the method comprising:
receiving visual input data captured by a stereo pair of cameras, the stereo pair of cameras being calibrated and the visual input data comprising a plurality of image frames; detecting points and lines in the received visual input data; finding point correspondences between the detected points and line correspondences between the detected lines in the visual input data; tracking the detected points and lines temporally across the plurality of the image frames; and estimating the camera motion using the detected points and lines in the visual input data, wherein estimating the camera motion comprises:
forming a first trifocal tensor and a second trifocal tensor, the first trifocal tensor comprising the calibrated stereo pair of the cameras and a first camera of a corresponding stereo pair of cameras and the second trifocal tensor comprising the calibrated stereo pair of the cameras and a second camera of the corresponding stereo pair of cameras;selecting a three-feature combination of detected points and lines, the three-feature combination being at least one of: a combination of two detected points and one detected line or a combination of one detected point and two detected lines; andestimating a plurality of camera motion parameters associated with the first trifocal tensor and the second trifocal tensor from the selected combination of detected points and lines.