Dynamic Full Body Pose Image Cropping for Hair Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hair analysis tools face challenges in accurately determining hair properties from full body pose images due to irrelevant information and varying lighting conditions, leading to poor segmentation of hair regions.
Innovation Solution
A method for dynamic cropping of full body pose images, involving face detection, distance approximation, and coordinates mapping to isolate the hair region, improving segmentation accuracy by removing irrelevant information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If full body pose images are used for hair analysis, then more comprehensive user information is captured, but irrelevant information increases and segmentation accuracy deteriorates
Solution Approach 1:
The patent applies segmentation by dividing the full body pose image into multiple regions: face region, hair region, and body region. By detecting the face bounding box and calculating distance metrics, the system dynamically determines the hair region boundaries and segments it from the rest of the image, thereby improving segmentation accuracy while maintaining comprehensive information capture
Solution Approach 2:
The patent extracts the relevant hair region from the full body pose image by calculating distance metrics from the face bounding box. The hair region is identified and extracted dynamically based on the detected face position and predetermined distance thresholds, separating useful information from irrelevant background information
2Ease of manufacture
If static cropping based on face detection is used, then processing simplicity is maintained, but dynamic adaptation to varying distances is lost and segmentation accuracy deteriorates
Solution Approach 1:
The patent implements dynamics by calculating the distance from the face bounding box to dynamically determine the hair region boundaries. Instead of using fixed cropping parameters, the system adjusts the region of interest based on the detected face position and predetermined distance metrics, enabling adaptive processing that maintains both simplicity and accuracy
3Measurement precision
If deep learning models are trained to identify hair regions, then segmentation capability is improved, but system complexity and training requirements increase
Solution Approach 1:
The patent uses the detected face bounding box as an intermediary to determine the hair region boundaries. By calculating distance metrics from the face position, the system indirectly identifies the hair region without requiring complex deep learning models trained specifically for hair detection, thereby reducing system complexity while maintaining segmentation accuracy
Data Source
AI summary
Disclosed is method for dynamic cropping of full body pose (FBP) image, method comprising: receiving in computing device FBP image from image resource, wherein FBP image comprises face; detecting in received FBP image area of face by bounding box; detecting coordinates of corner points of bounding box; performing distance approximation of detected area of face in received FBP image; cropping distance approximated area of face of FBP image; performing coordinates mapping transformation of cropped FBP image by mapping coordinates of cropped FBP image to received FBP image; segmenting region of hair of cropped FBP image; masking region of hair of cropped FBP image for obtaining hair polymask; and applying masked FBP image to received FBP image according to performed coordinates mapping for obtaining the dynamically cropped FBP image.