Optical Tracking with CMOS Windowing for Low-Latency Pose Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical tracking systems face challenges in achieving high-speed and low-latency data delivery for pose tracking applications, particularly in medical and industrial settings, due to the need for processing large volumes of data and the requirement for sterile, disposable markers.
Innovation Solution
The implementation of CMOS windowing and compression techniques within the optical tracking system, which involves partial data readout and compression to reduce data processing time, combined with the use of a micro-lens array for simultaneous 2D and 3D image capture, enables high-speed and low-latency data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional optical tracking systems process full sensor data, then measurement precision is maintained, but processing time increases and speed decreases
Solution Approach 1:
The patent segments the sensor data processing by dividing the image sensor data into multiple regions or frames, processing only relevant segments containing fiducial markers rather than the entire data set. This segmentation enables faster processing while maintaining tracking precision by focusing computational resources on critical areas.
Solution Approach 2:
The patent extracts and processes only the essential data elements needed for tracking - specifically the fiducial marker positions and movements - while discarding or deferring processing of redundant information. This extraction approach reduces processing time while preserving the measurement precision required for accurate pose determination.
2Productivity
If data compression is applied to sensor data, then data processing time is reduced, but measurement precision may deteriorate
Solution Approach 1:
The patent applies different processing qualities to different data regions - using full precision for fiducial marker data while applying compression or reduced processing to background or non-critical areas. This local quality approach maintains measurement precision for essential tracking data while improving overall processing efficiency through selective compression.
3Speed
If CMOS windowing is used to read partial data, then processing speed increases, but the quantity of information may be insufficient
Solution Approach 1:
The patent uses preliminary actions by predicting or pre-identifying the locations of fiducial markers in successive frames, allowing the CMOS sensor to window into only the relevant regions of interest. This preliminary identification ensures that partial data readout captures sufficient information for accurate tracking while maximizing readout speed.
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 data processing time and latency, allowing for faster and more precise tracking with reduced hardware costs, particularly in medical applications such as robotic surgery, by using compressed partial images and micro-lens arrays to enhance tracking speed and accuracy.
Implementation Method 1
an optical tracking system (10) that sense fiducials (22)... an optical sensor (11) providing sensor data
Implementation Method 2
uses an additional micro-lens array in the front of the image sensor to enable light field applications
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An optical tracking method is disclosed. The method comprises: generating, by at least one optical sensor, sensor data in response to detection of at least one signal, wherein the at least one signal comprises detection of at least one feature; extracting a portion of the sensor data based on one or more of a plurality of windows, wherein the extracted portion of the sensor data corresponds to a partial image; compressing at least the extracted portion of the sensor data to generate a compressed image; determining at least one of a position or an orientation of the feature from the compressed image; and transmitting at least one of the position or the orientation of the feature through a communication channel to an application.