Image-based position detection method, image based position detection apparatus, device and storage medium

Patent metadata

Field	Value
Publication number	US-11074709-B2
Application number	US-201916454910-A
Country	US
Kind code	B2
Filing date	Jun 27, 2019
Priority date	Jul 3, 2018
Publication date	Jul 27, 2021
Grant date	Jul 27, 2021

Abstract

Embodiments of the present application provide an image-based position detection method, an apparatus, a device and a storage medium. Images captured at the same time by a plurality of photographing devices mounted in different orientations are acquired, where the plurality of photographing devices are synchronous in time; a two-dimensional position of a target object in each of the images is detected; and a three-dimensional position of the target object in an actual space is determined based on the two-dimensional position of the target object in each of the images and internal parameters and external parameters of the plurality of photographing devices. The embodiments of the present application implement a three-dimensional positioning solution based on a plurality of cameras, thereby improving the reliability and accuracy of object positioning.

First claim

What is claimed is:

1. An image-based position detection method, comprising:
acquiring images captured at a same time by a plurality of photographing devices mounted in different orientations, wherein the plurality of photographing devices are synchronous in time;
detecting a two-dimensional position of a target object in each of the images;
determining a three-dimensional position of the target object in an actual space based on the two-dimensional position of the target object in each of the images and internal parameters and external parameters of the plurality of photographing devices; and
correcting the two-dimensional position of the target object in the images captured by each of the photographing devices based on the three-dimensional position of the target object in the actual space and the external parameter and the internal parameter of each of the photographing devices using a geometric theory.

2. The method according to claim 1 , wherein the detecting a two-dimensional position of a target object in each of the images comprises:
detecting the two-dimensional position of the target object in each of the images based on a preset detection model.

3. The method according to claim 2 , wherein the detecting the two-dimensional position of the target object in each of the images based on a preset detection model comprises:
detecting a two-dimensional position of each part of the target object in each of the images based on the preset detecting model; and
determining the two-dimensional position of the target object as a whole in each of the images based on the two-dimensional position of each part of the target object in each of the images.

4. The method according to claim 1 , wherein the images comprise depth images.

5. The method according to claim 4 , wherein before the detecting a two-dimensional position of a target object in each of the images, the method further comprises:
acquiring the depth images captured by the plurality of photographing devices when the target object is not in the actual space as reference images;
the detecting a two-dimensional position of a target object in each of the images comprises:
determining, for each of the photographing devices, the two-dimensional position of the target object in each of the images based on a difference between depth information of a depth image captured by the photographing device at the time of capture and depth information of a reference image captured by the photographing device.

6. The method according to claim 1 , wherein the method further comprises:
determining, based on the three-dimensional position of the target object in the actual space and the internal parameter and the external parameter of each of the photographing devices, the two-dimensional positions of the target object in images in which the target object is shielded by other objects.

7. The method according to claim 1 , wherein the internal parameter of each of the photographing devices comprises at least one of a focal length, a field of view and a resolution, and the external parameter of each of the photographing devices comprises at least one of a coordinate position, an orientation and a pitch angle.

8. A non-transitory computer readable storage medium, having a computer program stored thereon, wherein the program, when executed by a processor, implements the method according to claim 1 .

9. An image-based position detection apparatus, comprising:
one or more processors; and
a storage apparatus, configured to store one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to:
acquire images captured at a same time by a plurality of photographing devices mounted in different orientations, wherein the plurality of photographing devices are synchronous in time;
detect a two-dimensional position of a target object in each of the images;
determine a three-dimensional position of the target object in an actual space based on the two-dimensional position of the target object in each of the images and internal parameters and external parameters of the plurality of photographing devices; and
correct the two-dimensional position of the target object in the images captured by each of the photographing devices based on the three-dimensional position of the target object in the actual space and the external parameter and the internal parameter of each of the photographing devices using a geometric theory.

10. The apparatus according to claim 9 , wherein the one or more programs further cause the processor to:
detect the two-dimensional position of the target object in each of the images based on a preset detection model.

11. The apparatus according to claim 10 , wherein the one or more programs further cause the processor to:
detect a two-dimensional position of each part of the target object in each of the images based on the preset detection model; and
determine the two-dimensional position of the target object as a whole in each of the images based on the two-dimensional position of each part of the target object in each of the images.

12. The apparatus according to according to claim 9 , wherein the images comprise depth images.

13. The apparatus according to according to claim 12 , wherein the one or more programs further cause the processor to:
acquire the depth images captured by the plurality of photographing devices when the target object is not in the actual space as reference images; and
determine, for each of the photographing devices, the two-dimensional position of the target object in each of the images based on a difference between depth information of a depth image captured by the photographing device at the time of capture and depth information of a reference image captured by the photographing device.

14. The apparatus according to claim 9 , wherein the one or more programs further cause the processor to:
determine, based on the three-dimensional position of the target object in the actual space and the internal parameter and the external parameter of each of the photographing devices, the two-dimensional positions of the target object in images in which the target object is shielded by other objects.

15. The apparatus according to claim 9 , wherein the internal parameter of each of the photographing devices comprises at least one of a focal length, a field of view and a resolution, and the external parameter of each of the photographing devices comprises at least one of a coordinate position, an orientation and a pitch angle.

Assignees

• Baidu online network technology beijing co ltd

Inventors

• Li Xubin
• Wang Jian
• Wen Shilei
• Ding Errui
• Sun Hao
• Kang Le
• Bao Yingze
• Chen Mingyu
• Chi Zhizhen
• Liu Zeyu
• Zhang Chengyue

Classifications

• G06T2207/30196

  Human being; Person · CPC title

• G06T7/254 Primary

  involving subtraction of images · CPC title

• G06T7/70 Primary

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06T2207/10004

  Still image; Photographic image · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title