Wearable Camera Frame Control for Lower-Power Vision Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable devices with dedicated cameras for vision processing consume excessive power due to unnecessary image frame output, leading to inefficiency and increased size.
Innovation Solution
Implementing a method to control camera settings and frame rates based on the specific vision processing tasks, including dummy image frames and adjusting frame rates to minimize power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a wearable device includes dedicated cameras for respective vision processing processes, then the device can perform multiple vision processing tasks simultaneously, but the volume of parts increases making it difficult to reduce device size
Solution Approach 1:
The patent applies multi-functionality by enabling a single camera to perform multiple vision processing tasks (SLAM, head tracking, hand tracking, surface reconstruction) through dynamic camera setting adjustments. The camera can be reconfigured between different frame rates and exposure times depending on which vision processing process needs to be performed, eliminating the need for separate dedicated cameras for each function.
2Productivity
If the camera obtains image frames at high frame rate for different vision processing processes, then the device can maintain high performance in all vision tasks, but power consumption increases due to unnecessary image frame output
Solution Approach 1:
The patent applies dynamics by dynamically adjusting camera settings (frame rate, exposure time) based on which vision processing process is currently being performed. The system selectively changes camera operation parameters to match the specific requirements of each vision task, rather than maintaining fixed high-performance settings continuously. This dynamic adaptation reduces power consumption by lowering frame rates when full performance is not needed.
Solution Approach 2:
The patent changes physical parameters of camera operation, specifically frame rate and exposure time, to optimize power consumption. By adjusting these parameters according to the specific vision processing requirements, the system can maintain necessary performance while reducing unnecessary image frame capture and processing, thereby lowering overall power consumption.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A wearable apparatus including a camera and a method of controlling the same is provided. The method includes determining a vision processing process to be performed on each of image frames obtained at a first frame rate, determining a dummy image frame on which the vision processing process is not to be performed, applying, to a camera, a camera setting corresponding to a vision processing process to be performed on a current image frame, obtaining the current image frame according to the applied camera setting, outputting the obtained current image frame from the camera, and setting an exposure time of the camera setting to be less than or equal to an threshold value or changing the first frame rate to a second frame rate in response to the current image frame being the dummy image frame.