Green Screen Foreground Extraction for Driven Target Parts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reusing a pre-segmentation mask to segment a driven image in green screen image processing often leads to inaccuracies in the foreground extraction, resulting in poor segmentation effects due to misalignment of the target object's region in the driven image.
Innovation Solution
Fusing a pre-segmentation mask with a green screen image to create a composite image, determining a target region, driving the target part, and extracting the driven target part using pixels of the driven image and background region, rather than reusing the pre-segmentation mask directly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a pre-segmentation mask is reused to segment a driven image, then the processing efficiency is improved, but the segmentation accuracy deteriorates due to region misalignment
Solution Approach 1:
The method performs preliminary action by fusing the pre-segmentation mask with the green screen image to create a composite image that integrates both the original segmentation information and the green screen data. This preliminary fusion prepares the data structure for subsequent accurate extraction of the driven target part, allowing the system to maintain efficiency while improving accuracy in the final segmentation step
Solution Approach 2:
The composite image serves as an intermediary that bridges the pre-segmentation mask and the final driven image segmentation. By creating this intermediate representation that combines mask information with green screen image data, the system can accurately extract the driven target part without directly reusing the potentially misaligned pre-segmentation mask, thus resolving the accuracy-efficiency contradiction
2Device complexity
If a pre-segmentation mask is reused for segmentation, then the computational complexity is reduced, but the segmentation quality deteriorates due to inclusion of background pixels and missing target pixels
Solution Approach 1:
The method extracts only the necessary components from the composite image and green screen image to create the driven target part segmentation. Instead of directly reusing the entire pre-segmentation mask, the system extracts specific pixel information from the driven image and green screen image, obtaining a precise segmentation that excludes background pixels and includes all target pixels without increasing overall computational complexity
Solution Approach 2:
The method changes the parameters used for segmentation by switching from directly reusing the pre-segmentation mask parameters to using pixel-based extraction from the composite image and green screen image. This parameter change allows the system to adapt to the driven image's actual content, improving segmentation quality while maintaining computational efficiency through targeted pixel processing
Data Source
AI summary
An image processing method and apparatus, an electronic device, and a storage medium are provided. The method includes fusing a pre-segmentation mask corresponding to a green screen image and the green screen image to obtain a composite image comprising a foreground region and a background region, the foreground region having a target object, and the target object comprising a target part; determining a target region comprising the target part in the composite image; driving the target part in the target region; determining a target region comprising the driven target part as a driven image; extracting the driven target part in the driven image according to pixels of the driven image and pixels of the background region; and generating a target foreground region corresponding to the target object according to the driven target part and a region other than the target part in the foreground region.