Region-Specific Image Distortion Correction for Natural Wide-Angle Portraits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image distortion correction methods fail to naturally correct portrait and background distortions in images captured by terminals with wide-angle lenses, leading to discontinuities and poor user experience.
Innovation Solution
An image distortion correction method that applies different projection algorithms to distinct regions of a portrait (head, body, and background) and the field of view edge, ensuring natural and continuous transitions while minimizing field of view loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a homography constraint is established for the rectangular object region to correct stretching deformation, then the rectangular object region can be corrected, but the portrait cannot be corrected in detail and discontinuity occurs in the background region
Solution Approach 1:
The patent divides the image into multiple regions including rectangular object regions and non-rectangular object regions, and applies different correction strategies to each region. This segmentation allows the rectangular regions to be corrected using homography constraints while non-rectangular regions are handled separately to maintain overall image continuity and avoid discontinuities in the background.
Solution Approach 2:
The patent applies different correction methods to different regions of the image based on their specific characteristics. Rectangular regions use homography transformation while non-rectangular regions use other correction approaches, ensuring that each region is corrected according to its local properties rather than applying a uniform correction method to the entire image.
2Manufacturing precision
If straight line detection algorithm is used to determine straight line region in background, then straight line constraint can be established, but the final image correction effect is affected by straight line detection precision
Solution Approach 1:
The patent performs preliminary region division and identification before applying correction constraints. By pre-segmenting the image into different object regions and background areas, the system establishes correction constraints based on these pre-defined regions rather than relying on complex real-time straight line detection, thereby reducing computational complexity while maintaining correction precision.
3Ease of manufacture
If a portrait is corrected as a whole, then the correction process is simple, but the final portrait does not really fit human vision and natural appearance
Solution Approach 1:
The patent segments the portrait into different body regions such as head, torso, and limbs, and applies region-specific correction parameters to each segment. This allows the correction process to maintain relative simplicity while achieving natural appearance by adjusting each body part according to its specific geometric characteristics and expected visual appearance.
Solution Approach 2:
The patent applies different correction strategies to different parts of the portrait based on local anatomical characteristics. For example, the head region may use different correction parameters than the limb regions, ensuring that each part is corrected to match human visual expectations while maintaining overall process efficiency.
Data Source
Figure 1(1)~1(2)
Figure 2
Figure 3
AI summary
This application provides an image distortion correction method and an apparatus. The method includes: performing optical distortion correction on a collected source image, to obtain a first corrected image, where the first corrected image includes a background region and a portrait in which stretching deformation occurs, and the portrait in which stretching deformation occurs includes at least a first human body region in which stretching deformation occurs and a second human body region in which stretching deformation occurs; and performing algorithm constraint correction on the first human body region, the second human body region, and the background region, to obtain a second corrected image, where constraint terms used for the algorithm constraint correction respectively constrain the first human body region, the second human body region, and the background region. In this way, the stretching deformation in both the first human body region and the second human body region is corrected, and for human vision there is no discontinuity in image content throughout the portrait in the background region. In this application, natural image distortion correction can be implemented, and therefore user experience is improved.