Do Not Capture Face Obfuscation for Privacy Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The pervasive use of mobile devices and wearable cameras can compromise the privacy of individuals by capturing their images without consent, leading to potential publication on social media and the internet, making it difficult to control privacy leakage in third-party and public photos and videos.
Innovation Solution
The Do Not Capture (DNC) technology uses face detection and matching algorithms, combined with motion and orientation information, to obfuscate unwilling subjects in captured media by blurring or removing their images before they are saved or shared, operating efficiently on mobile devices without the need for cloud support.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face detection and matching algorithms are used to identify subjects in captured media, then privacy protection accuracy is improved, but device complexity and processing power requirements increase
Solution Approach 1:
The system performs preliminary face detection and tracking before the actual capture moment, establishing subject identification in advance. Motion information and trajectory data are collected and analyzed beforehand to predict subject positions, enabling faster and more accurate obfuscation decisions during capture without requiring complex real-time processing of all captured data.
Solution Approach 2:
Motion information serves as an intermediary element that bridges face detection and final obfuscation decisions. By incorporating trajectory analysis and motion patterns as intermediate processing layers, the system achieves higher privacy protection accuracy while distributing computational complexity across multiple processing stages rather than requiring all computations to occur simultaneously in a single complex system.
2Reliability
If real-time obfuscation is performed on captured media, then privacy protection effectiveness is improved, but energy consumption and battery usage increase
Solution Approach 1:
The system employs periodic face detection and obfuscation operations rather than continuous processing. By detecting faces at specific intervals and only processing frames where subjects are actually present, the system maintains effective privacy protection while significantly reducing overall energy consumption compared to continuous real-time processing of all captured media.
Solution Approach 2:
The system extracts and processes only the essential elements needed for privacy protection—specifically face regions and motion trajectories—rather than processing entire captured media files. By isolating and obfuscating only the relevant subject portions, the system achieves effective privacy protection with reduced computational load and lower energy consumption.
3Measurement precision
If motion information and trajectory matching are used to identify subjects, then subject identification accuracy is improved, but data processing requirements and computational load increase
Solution Approach 1:
The system implements partial processing of motion information by focusing only on key trajectory parameters and dominant motion patterns rather than analyzing complete motion data sets. This selective approach maintains high subject identification accuracy by capturing essential motion characteristics while reducing computational load through simplified processing of motion vectors and trajectory points.
Data Source
AI summary
Obfuscating a human or other subject in digital media preserves privacy. A user of a smartphone, for example, may enable a flag for obscuring her face in digital photos or movies. When any device captures digital media, the user's smartphone transmits the flag for receipt. The device capturing the digital media is thus informed of the user's desire to obscure her face or even entire image. The device capturing the digital media may thus perform an obscuration in response to the flag.