Visual localization method and apparatus
AssigneeHuawei Technologies Co., Ltd.
InventorsWensen Feng, Huan Zhang, Jun Cao, Jiange Ge, Zhongwei Tang, Jiangwei Li
Life of this patent
Estimating the term…
Abstract
A visual localization method includes obtaining an image captured by a terminal device; obtaining two-dimensional line feature information of the image that includes at least one of information about a boundary between a building and a non-building or information about a boundary between a non-building and a non-building; and determining a localization pose of the terminal device based on location information of the terminal device, magnetometer angle deflection information of the terminal device, a satellite map, and the two-dimensional line feature information.
Claims19
A visual localization method implemented by a server, the visual localization method comprising:
- obtaining an image from a terminal device;
- obtaining two-dimensional line feature information of the image, the two-dimensional line feature information comprising at least one of first boundary information about a first boundary between a building and a first non-building, or second boundary information about a second boundary between a second non-building and a third non-building;
- determining a candidate pose set based on first location information of the terminal device and magnetometer angle deflection information of the terminal device;
- determining N optimized poses based on the candidate pose set, the two-dimensional line feature information, and a satellite map, wherein N is an integer greater than 1;
- and
- determining a localization pose of the terminal device based on the N optimized poses.
A visual localization apparatus comprising:
- a memory configured to store instructions;
- and
- a processor coupled to the memory and configured to execute the instructions to cause the visual localization apparatus to:
- obtain an image captured by a terminal device;
- obtain two-dimensional line feature information of the image, the two-dimensional line feature information comprising at least one of first boundary information about a first boundary between a building and a first non-building, or second boundary information about a second boundary between a second non-building and a third non-building;
- determine a candidate pose set based on first location information of the terminal device and magnetometer angle deflection information of the terminal device;
- determine N optimized poses based on the candidate pose set, the two-dimensional line feature information, and a satellite map, wherein N is an integer greater than 1;
- and
- determine a localization pose of the terminal device based on the N optimized poses.
A computer program product comprising computer-executable instructions that are stored on a non-transitory computer storage medium and that, when executed by a processor, cause a visual localization apparatus to:
- obtain an image captured by a terminal device;
- obtain two-dimensional line feature information of the image, the two-dimensional line feature information comprising at least one of first boundary information about a first boundary between a building and a first non-building, or second boundary information about a second boundary between a second non-building and a third non-building;
- determine a candidate pose set based on first location information of the terminal device and magnetometer angle deflection information of the terminal device;
- determine N optimized poses based on the candidate pose set, the two-dimensional line feature information, and a satellite map, wherein N is an integer greater than 1;
- and
- determine a localization pose of the terminal device based on the N optimized poses.
Dependencies are inferred from the claim text.