Parallel Eye Detection Units for Low Latency Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current eye detection systems suffer from high latency, inefficient power consumption, and limited frame rates, making them unsuitable for real-time applications, particularly in computer-based systems that require enhanced user interaction and monitoring.
Innovation Solution
An eye detection system that processes subsets of digital image frames in parallel using multiple eye detection units, each handling a specified region of the image, with an eye feature extraction module and an eye data generating module that only generates data when predetermined eye features are identified, allowing for reduced latency and power consumption while increasing frame rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If eye detection systems process images sequentially one frame at a time, then the system complexity is reduced, but the frame rate is limited and latency increases
Solution Approach 1:
The patent divides the image processing task into multiple segments by splitting the image into multiple regions of interest (ROIs), each processed by a separate eye detection unit. This segmentation enables parallel processing of different image regions simultaneously, increasing frame rate while keeping each individual processing unit relatively simple.
Solution Approach 2:
The patent transitions from sequential single-threaded processing to parallel multi-threaded processing by creating multiple eye detection units that operate simultaneously on different ROIs. This dimensional change in processing architecture allows multiple operations to occur concurrently, dramatically improving frame rate.
2Measurement precision
If eye detection systems analyze the entire image frame, then detection accuracy is maintained, but power consumption increases
Solution Approach 1:
The patent extracts and processes only the relevant portions of the image (regions of interest containing eyes) rather than analyzing the entire image frame. By taking out and processing only the necessary ROIs, the system maintains eye detection accuracy while significantly reducing the computational load and power consumption.
Solution Approach 2:
The patent applies partial action by processing only a subset of the image data (specific ROIs) rather than the complete image frame. This selective processing approach maintains sufficient detection accuracy for the application while reducing overall computational effort and power consumption.
3Reliability
If eye detection systems use dedicated computer hardware, then processing reliability is improved, but the system size and cost increase
Solution Approach 1:
The patent implements eye detection functionality within the existing GPU architecture, making the GPU serve multiple functions (graphics rendering and eye detection). This multi-functionality approach maintains processing reliability by using a dedicated hardware resource while avoiding the need for additional dedicated computer hardware, thus reducing system size.
Solution Approach 2:
The patent merges the eye detection processing functions with the existing GPU hardware resources. By combining these functions into a single integrated system rather than using separate dedicated hardware, the patent maintains reliability through hardware-based processing while reducing overall system size and cost.
4Speed
If eye detection systems process high frame rates, then real-time performance is improved, but latency between user action and response increases due to frame-by-frame handling
Solution Approach 1:
The patent performs preliminary processing by pre-defining and preparing multiple regions of interest before actual eye detection occurs. This preliminary setup enables the system to quickly process incoming frames by directly analyzing predetermined ROIs without needing to search the entire frame, thereby reducing response latency while maintaining high frame rates.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly reduces latency and power consumption while enhancing frame rates, enabling more efficient and effective eye detection in computer-based systems, including laptops and other devices, by processing image data in parallel and only analyzing regions with potential eye features.
Implementation Method 1
at least one illuminator element adapted to send out light in a direction towards the head of a user
Implementation Method 2
at least one image sensor being adapted to receive light from the illuminated head of the user and repeatedly capture pictures
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The invention generally relates to systems and methods for eye detection for use in, for example, installations and systems for tracking eyes and gaze angle/directions. The system includes a plurality of eye detection units connected to an image sensor. The eye detection units are adapted to receive a respective data subset including picture elements of a digital image frame relating to a respective pre-selected region of the image sensor substantially in parallel and wherein the eye detection units are adapted to operate substantially in parallel, each eye detection unit being adapted to receive a respective data subset and including an eye feature extraction module adapted to perform an eye feature extraction process on the respective data subset to identify predetermined eye features. An eye data generating module is adapted to receive information of potential identified eye features including picture element coordinates for the identified eye features, and the received data subsets, and generate eye data based on the information of potential eye features, wherein the eye data is generated if identified eye features fulfil predetermined conditions, and transfer generated eye data and the received regions of picture elements to a processing module, and wherein a data flow of the eye data and the data subsets are substantially parallel.