Image registration method, device, terminal and storage medium
By registering multiple target images using reference images with depth information, the problem that the prior art is difficult to complete image registration in extreme scenarios is solved, and efficient image registration with simple and low calculation amount is achieved, and the shooting effect is improved.
CN114723788BActive Publication Date: 2025-03-18BEIJING XIAOMI MOBILE SOFTWARE CO LTD
1 Cites 0 Cited by
Patent Information
- Application Number
- CN202110001686.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-01-04
- Publication Date
- 2025-03-18
- Estimated Expiration
- 2041-01-04
AI Technical Summary
Technical Problem
The prior art is difficult to complete the registration of multiple images in extreme scenarios, affecting subsequent processing and shooting effects.
Method used
By introducing a reference image with depth information, multiple target images are registered using the depth information of the reference image, including determining a point cloud image based on the world coordinate system of the reference image, and registering the target image according to the point cloud image and mapping relationship.
Benefits of technology
The algorithm is simplified, the calculation amount is reduced, and it is suitable for a variety of extreme scenes. It can effectively complete the registration of multiple images, facilitate subsequent processing and improve shooting effect.
✦ Generated by Eureka AI based on patent content.
Abstract
The present disclosure relates to an image registration method, device, terminal and storage medium, the method comprising: obtaining a reference image of a photographed object and multiple target images; wherein the target image and the reference image are images generated by a terminal photographing the same photographed object at the same photographing position, and the reference image includes depth information; and performing image registration on the multiple target images according to the depth information of the reference image. The algorithm of the method is simple, the amount of calculation is small, and it is applicable to a variety of extreme scenes (such as scenes with very dark light, scenes with many repeated features of the photographed object, scenes with unclear features of the photographed object, or scenes with large information differences of multiple images obtained by multiple camera modules, etc.), and can well complete the registration of multiple target images, facilitate the subsequent processing of multiple target images, and thus improve the shooting effect.
Need to check novelty before this filing date? Find Prior Art