Method and apparatus with estimation of distance between pedestrian and camera
A method with distance estimation includes detecting a pedestrian region of a pedestrian comprised in a plurality of images received from a camera; determining a static point in the detected pedestrian region; and determining a distance between the pedestrian and the camera based on the static point in each of the images and a position of the camera corresponding to each of the images.
1. A method with distance estimation, the method comprising:
detecting a pedestrian region of a pedestrian comprised in a plurality of images obtained from a camera at different positions;
determining a static point in the detected pedestrian region, including extracting a plurality of joint regions of the pedestrian from the pedestrian region and determining the static point of the pedestrian based on the extracted joint regions;
determining 3D (Three-dimensional) points based on a vector associated with the extracted joint regions and determining 3D coordinates of each of the extracted joint regions; and
determining a distance between the pedestrian and the camera, using the 3D points and the 3D coordinates, based on a difference between the static point in each of the images and the different positions of the camera corresponding to each of the images.
2. The method of claim 1 , wherein the static point corresponds to a position of a stepping foot of the pedestrian.
3. The method of claim 1 , wherein the determining of the static point comprises:
determining, to be the static point, a point corresponding to a position of a stepping foot of the pedestrian in the joint region.
4. The method of claim 1 , wherein the determining of the static point comprises:
determining, to be the static point, a point at which a height is lowest in the joint regions.
5. The method of claim 1 , wherein the determining of the distance between the pedestrian and the camera comprises:
determining the distance between the pedestrian and the camera based on a difference between three-dimensional (3D) points determined by a vector comprising the static point in each of the images and a point corresponding to a position of the different positions of the camera.
6. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the method of claim 1 .
7. An apparatus with distance estimation, the apparatus comprising:
a processor configured to:
detect a pedestrian region of a pedestrian comprised in a plurality of images obtained from a camera at different positions;
determine a static point in the detected pedestrian region, including extracting a plurality of joint regions of the pedestrian from the pedestrian region and determining the static point of the pedestrian based on the extracted joint regions;
determine 3D (Three-dimensional) points based on a vector associated with the extracted joint regions, and determine 3D coordinates of each of the extracted joint regions; and
determine a distance between the pedestrian and the camera, using the 3D points and the 3D coordinates, based on a difference between the static point in each of the images and the different positions of the camera corresponding to each of the images.
8. The apparatus of claim 7 , wherein the static point corresponds to a position of a stepping foot of the pedestrian.
9. The apparatus of claim 7 , wherein, for the determining of the static point, the processor is configured to:
determine, to be the static point, a point corresponding to a position of a stepping foot of the pedestrian in the joint regions.
10. The apparatus of claim 7 , wherein, for the determining of the static point, the processor is configured to:
determine, to be the static point, a point at which a height is lowest in the joint regions.
11. The apparatus of claim 7 , wherein, for the determining of the distance between the pedestrian and the camera, the processor is configured to:
determine the distance between the pedestrian and the camera based on a difference between three-dimensional (3D) points determined by a vector comprising the static point in each of the images and a point corresponding to a position of the different positions of the camera.
12. The apparatus of claim 7 , further comprising the camera, wherein the camera is configured to collect the images.
13. A method with distance estimation, the method comprising:
determining a static point of a target in each of images obtained by a camera at different positions, including extracting a plurality of joint regions of the target from a target region, included in each of the images, and determining the static point of the target based on the extracted joint regions;
determining 3D (Three-dimensional) points based on a vector associated with the extracted joint regions and determining 3D coordinates of each of the extracted joint regions; and
determining a distance between the target and the camera, using the 3D points and the 3D coordinates, based on the static point in each of the images and the different positions of the camera.
14. The method of claim 13 , wherein, for each of the images, the determining of the static point comprises determining, to be the static point, a lowest point among points in the joint regions.
15. The method of claim 13 , wherein
the camera obtains the images at the different positions by a movement of the camera, and
the determining of the distance between the target and the camera comprises:
determining a structure from motion (SFM) based on the static point in each of the images and the different positions corresponding to the images; and
determining the distance based on the SFM.