Head Region Editing via Facial Expression Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods that rely on manual user input for determining head regions in images are inaccurate, leading to inadequate editing effects due to the inclusion of surrounding regions.
Innovation Solution
An image processing method that utilizes a trained pixel classification model to classify each pixel in an image and determine the head region with precision, enabling pixel-level head recognition and fine matting of head edges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual rectangular box determination is used to select head region, then operation simplicity is improved, but head region identification accuracy deteriorates due to inclusion of surrounding regions
Solution Approach 1:
The system automatically determines the head region by detecting facial features and classifying pixels, eliminating the need for manual rectangular box selection. The processor autonomously identifies the head region based on facial feature detection results, allowing the system to serve itself rather than requiring user input for region selection.
Solution Approach 2:
The patent replaces the manual mechanical operation (swiping or dragging to define a rectangular box) with an automated computational approach using facial feature detection and pixel classification algorithms. This substitution eliminates the need for manual region selection while achieving more precise head region identification.
2Device complexity
If manual rectangular box determination is used, then device complexity is reduced, but head region segmentation precision deteriorates
Solution Approach 1:
The patent segments the image into multiple pixel regions and classifies each pixel based on its characteristics and spatial relationships with detected facial features. This pixel-level segmentation enables precise head region identification while maintaining manageable system complexity through structured processing steps.
Solution Approach 2:
The system transitions from coarse rectangular region selection to fine-grained pixel-level classification, adding a dimensional level of precision. By operating at the pixel dimension rather than the region dimension, the system achieves superior segmentation precision without proportionally increasing overall system complexity.
Data Source
AI summary
Aspects of the disclosure provide an image processing method, an image processing terminal, an image processing apparatus, and a non-transitory computer-readable storage medium. The method can include performing face detection on a target image to detect a face region in the target image. The method includes determining an expression class of the detected face region based on a trained expression recognition model and determining whether the determined expression class of the detected face region is a target expression class. If the determined expression class is the target expression class, head detection is performed to identify a head region in the target image and the identified head region is edited. If the determined expression class is not the target expression class, the head detection is not performed on the target image.