Facial Key-Point Detection for Low-Resource Face Bounding Boxes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional face detection techniques that rely on neural networks are computationally intensive, making them challenging to implement on low-budget devices, which often lack the necessary processing power, memory, and power resources.
Innovation Solution
An electronic device uses facial key-points and basic equations of computational geometry to detect faces, determining a bounding box without the high computational complexity of neural networks, enabling efficient face detection on resource-constrained devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If neural network-based face detection techniques are used, then face detection accuracy is improved, but processing power requirements and device complexity increase
Solution Approach 1:
The patent extracts only the essential facial key-points (eyes, eyebrows, nose, mouth, ears) from the complete face detection problem, rather than using full neural networks. This extraction approach maintains sufficient detection accuracy while dramatically reducing computational complexity and processing power requirements, enabling implementation on low-budget devices.
Solution Approach 2:
The patent replaces expensive, computationally intensive neural network models with simpler, more economical algorithms based on facial key-point detection. This substitution uses less powerful processing resources and can be executed efficiently on devices with limited computational capabilities, effectively trading off some complexity for broader device compatibility.
2Measurement precision
If neural network-based face detection techniques are used, then face detection accuracy is improved, but implementation difficulty on low-budget devices increases
Solution Approach 1:
The patent extracts only the essential facial key-points (eyes, eyebrows, nose, mouth, ears) from the complete face detection problem, rather than using full neural networks. This extraction approach maintains sufficient detection accuracy while dramatically reducing computational complexity and processing power requirements, enabling implementation on low-budget devices.
Solution Approach 2:
The patent replaces complex neural network mechanical systems with simpler geometric and algebraic calculations. By substituting the neural network processing mechanism with basic mathematical operations on detected key-points, the implementation becomes significantly easier to deploy on devices with limited resources.
3Speed
If facial key-point detection with computational geometry equations is used, then processing speed is improved, but face detection accuracy may be reduced
Solution Approach 1:
The patent extracts only the most critical facial key-points (eyes, eyebrows, nose, mouth, ears) necessary for accurate face detection. By focusing computational effort on these essential features rather than processing entire images through neural networks, the system achieves both high processing speed and sufficient detection accuracy.
Solution Approach 2:
The patent changes the parameter representation from full image data processed by neural networks to specific geometric parameters of facial key-points. This parameter transformation enables faster computational geometry-based processing while maintaining the essential information needed for accurate face detection and bounding box determination.
Data Source
AI summary
An electronic device and method for face detection based on facial key-points is provided. The electronic device receives an image of an object of interest. The image may include a face of the object of interest. The electronic device detects a plurality of key-points associated with the face of the object of interest based on the received image and determines a first coordinate value in the received image based on the detected plurality of key-points. Thereafter, the electronic device determines a region in the received image that includes the face of the object of interest based on the determined first coordinate value and controls the display device to overlay a marker onto the determined region of the face. The marker indicates a location of the determined region in the image.