Image Processing Reference Map Reuse for Chroma Key Artifact Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional chroma keying processes face inefficiencies due to imperfections in background materials, such as non-uniform colors and shadows, which are difficult to identify and remove, leading to low-quality overlays and computationally intensive processing.
Innovation Solution
A content editor uses a reference map created from a first image to mark foreground and background regions, allowing for efficient identification and removal of artifacts in subsequent images by comparing coloration and applying the map to subsequent frames, reducing the need for heavy processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional chroma keying processes are used to remove background artifacts, then background imperfections can be removed, but the processing becomes computationally intensive and time-consuming
Solution Approach 1:
The patent creates a reference map from a first image that identifies foreground and background regions before processing subsequent images. This preliminary action allows the system to quickly classify regions in subsequent frames without performing full artifact detection algorithms, thereby maintaining high removal quality while significantly improving processing speed.
Solution Approach 2:
The patent copies the foreground/background classification information from the reference map created from the first image and applies it to subsequent images. By reusing this reference information across multiple frames, the system avoids redundant computational operations while maintaining consistent artifact removal quality throughout the video sequence.
2Productivity
If a reference map from the first image is reused for subsequent images, then processing efficiency is improved, but accuracy may decrease for images with significant color variations
Solution Approach 1:
The patent dynamically adjusts the threshold range for coloration comparison based on the specific image being processed. When significant color variations are detected between the reference image and subsequent images, the system adapts by widening the threshold range or recalibrating the reference map, thereby maintaining both processing efficiency and coloration matching accuracy across diverse imaging conditions.
Data Source
AI summary
A content editor application receives a reference map that indicates which portions of a first image are foreground, and which portions of the first image are background. The content editor compares a coloration of regions in the first image to a coloration of regions in the second image. For regions in the second image that match a coloration of corresponding regions in the first image, or that are within a threshold range of coloration, the content editor uses the reference map to mark regions of the second image that are foreground and to mark which regions of the second image are background. Accordingly, the reference map of the first image can be used to identify whether regions of a second image or subsequent images in a sequence are foreground and which are background.