Adaptive Dewarping Wide Angle Video Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wide angle camera images, particularly those of human faces, suffer from distortion due to the projection of three-dimensional environments onto two-dimensional images, which can be distracting in video calls.
Innovation Solution
An adaptive dewarping method that detects a region of interest, such as a human face, and applies a dewarping function with input parameters to maintain the relative proportions of features, minimizing distortion by predicting a target vertical position and determining a tilt parameter for the dewarping process, effectively using a Mercator projection to map pixels and preserve conformity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of moving object
If a wide angle lens is used to capture a broad field of view, then the field of view spans a range of angles, but distortion occurs in the projected two-dimensional image particularly affecting human faces
Solution Approach 1:
The patent divides the video frame into multiple regions, identifying a region of interest (such as a human face) and applying different processing to this region versus the rest of the frame. The dewarping function is selectively applied to the crop region containing the object of interest, preserving its proportions while maintaining the wide angle effect in other areas.
Solution Approach 2:
The patent applies local dewarping to specific regions rather than uniformly processing the entire image. By detecting the crop region and generating input parameters based on this region's characteristics, the system preserves conformity and relative proportions of features within the region of interest while allowing other regions to maintain the original wide angle perspective.
2Shape
If uniform dewarping is applied to the entire video frame, then distortion is reduced across all regions, but the overall wide angle perspective and field of view are compromised
Solution Approach 1:
The patent segments the image processing into a detected crop region and the remainder of the frame. The dewarping function with specifically generated input parameters is applied only to the crop region containing the object of interest, while other regions retain the original wide angle characteristics, thus preserving both local conformity and global field of view.
Solution Approach 2:
The system applies different quality characteristics to different regions: the crop region receives dewarped processing with enhanced conformity and natural proportions, while the surrounding areas maintain the original wide angle distortion characteristics, achieving local improvement without global compromise.
3Ease of manufacture
If the dewarping function is applied with fixed parameters, then processing is simple and fast, but the object of interest may not maintain proper proportions when its position changes in the frame
Solution Approach 1:
The patent makes the dewarping parameters dynamic by generating them based on the detected crop region characteristics. The input parameters to the dewarping function are determined from the region of interest's position and properties, allowing the system to adapt to changing object positions while maintaining proper proportions. This dynamic parameter generation resolves the contradiction between processing simplicity and adaptability.
Solution Approach 2:
The system uses feedback from the detected crop region to determine the input parameters for dewarping. By analyzing the region of interest's characteristics and using this information to generate appropriate parameters, the system adapts to different object positions and maintains conformity, bridging the gap between simple fixed-parameter processing and complex adaptive processing.
Data Source
AI summary
A video system processes video frames from a wide angle camera to dewarp the video frames in a manner that preserves conformity of an object of interest. A crop region of a video frame corresponding to the object of interest is determined. An input parameter to a dewarping function is generated based on the detected crop region. The dewarping function is applied to the crop region using the input parameter to generate a dewarped video frame and the dewarped video frame is outputted. The input parameter may be generated in a manner that causes the dewarped video frame to have higher conformity and lower distortion in the region around the object of interest than in a region distant from the object of interest.