Edge-Based Face Recognition for Media Playback Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing media devices face challenges in identifying individuals in video content due to slow and expensive cloud-based facial recognition processes that may also be inaccessible.
Innovation Solution
Implementing an edge-based face recognition process on media devices using systems-on-a-chip (SoC) for local processing, which includes face detection and recognition technologies, allowing for local analysis of video frames and reducing the need for remote processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cloud-based facial recognition service is used, then facial recognition functionality is provided, but processing speed is slow and cost is high
Solution Approach 1:
The patent introduces an edge device as an intermediary between the media device and cloud service. The edge device performs facial recognition processing locally, acting as a mediator that eliminates the need to upload video data to the cloud, thereby reducing processing time while maintaining recognition functionality.
Solution Approach 2:
The patent transitions from a single-dimension cloud-based processing model to a multi-dimensional architecture involving media device, edge device, and cloud service. By adding the edge device dimension, the system achieves local processing capabilities without sacrificing the option to use cloud services when needed.
2Reliability
If cloud-based facial recognition service is used, then facial recognition functionality is provided, but service accessibility is limited
Solution Approach 1:
The edge device serves as a local intermediary that enables facial recognition to function independently of cloud service availability. This mediator allows the system to operate in offline modes, improving accessibility and eliminating dependencies on remote cloud infrastructure.
3Productivity
If edge-based face recognition is implemented, then processing speed is improved, but device complexity increases
Solution Approach 1:
The patent segments the facial recognition system into distinct functional components: media device for video playback, edge device for local processing, and cloud service for supplementary functions. This segmentation allows each component to be optimized independently, managing overall system complexity while achieving fast local processing.
Solution Approach 2:
The edge device is designed with multi-functionality, capable of performing facial recognition, video processing, and acting as a local server. This universality reduces the need for separate dedicated devices, thereby managing complexity while maintaining processing capabilities.
Data Source
AI summary
A method includes accessing a video comprising a set of frames; receiving a command to identify a target displayed in a first frame; generating a first target data based on the target; generating a first confidence level between the first target data and a second target data stored in a remote database, wherein the second target data comprises identity data; in response to the first confidence level being at or above a confidence level threshold, outputting the identity data of the second target data; otherwise, generating a third target data based on the target displayed in a second frame of the set of frames; generating a second confidence level of similarity between the third target data and the second target data stored in the remote database; in response to the second confidence level being at or above the confidence level threshold, outputting the identity data of the second target data.