Portrait Creation Using Segmented Face and Hair Components
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing portrait creation devices face challenges in creating realistic portraits that balance detail and uniqueness, as portraits described by only outline shapes lack realism, while fine structures make them too intricate, and the lack of suitable part objects can result in unrecognizable facial images.
Innovation Solution
An image creating device and method that acquire and process images to extract feature information, specify hair style images based on facial features, and combine these with face component images to create a portrait, allowing for more expressive and detailed representations of facial features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the portrait is described by lines of only outline shapes of characteristic configurations of the face, then the portrait creation process is simple, but the portrait becomes poor in reality
Solution Approach 1:
The portrait creation process is divided into two independent stages: first creating a face component image with outline shapes of characteristic configurations, then separately creating a hair style image that is superimposed on the face component image. This segmentation allows each stage to focus on specific aspects without compromising overall quality.
Solution Approach 2:
The hair style image is extracted and created separately from the face component image. By taking out the hair style creation as an independent process, the system can incorporate detailed hair structures without overcomplicating the entire portrait creation process, thereby maintaining realism while preserving simplicity.
2Manufacturing precision
If the portrait is described by lines to the extent of fine structures of the face, then the touch and texture of the portrait become fine, but the unique quality of the portrait is decreased
Solution Approach 1:
The portrait is segmented into face component image and hair style image, with the hair style image containing the fine structures and textures. This segmentation concentrates the detailed work in one component, maintaining precision while avoiding excessive complexity in the overall structure.
Solution Approach 2:
Fine structures and textures are applied locally to the hair style image rather than throughout the entire portrait. This allows high manufacturing precision in the hair regions while keeping the face component image simpler, thereby maintaining unique quality while achieving detailed representation where needed.
3Productivity
If part objects are prepared for each region in advance, then the portrait creation is efficient, but resemblant part objects to actual face parts are sometimes not prepared
Solution Approach 1:
The hair style image serves multiple functions: it provides the hairstyle appearance, adds fine textures and touches, and can be adapted to match various facial features through the association with extracted feature information. This multi-functionality allows a single component to address multiple requirements without needing separate part objects for each function.
Solution Approach 2:
The system extracts feature information from the captured image and uses this feedback to specify and select appropriate hair style images from a plurality of candidates. This feedback mechanism ensures that the selected hair style image matches the actual facial features, thereby maintaining recognizability while preserving creation efficiency.
Data Source
AI summary
An image creating device includes: an acquiring unit, a component creation unit, an extraction unit, a specifying unit, and an image creation unit. The acquiring unit acquires an image. The component creation unit creates a face component image related to main components of a face included in the image acquired by the acquiring unit. The extraction unit extracts feature information from the face included in the image acquired by the acquiring unit. The specifying unit specifies a hair style image, which corresponds to the feature information extracted by the extraction unit, from hair style images which show outlines of hair and are associated with feature information of faces, the feature information being recorded in a recording section. The image creation unit creates a portrait image by using the hair style image specified by the specifying unit and the face component image created by the component creation unit.