Aliased Time-of-Flight Depth Segmentation for Power-Constrained Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optical time-of-flight (ToF) cameras require multiple exposures to achieve depth imaging, leading to high power consumption and limited frame rates, especially in battery-powered devices, and de-aliasing introduces motion artifacts and inefficiencies.
Innovation Solution
A method using aliased depth-image data for segmentation and tracking, employing a depth-imaging controller with a shutter-acquisition engine and segmentation engine to label coordinates in optical ToF cameras, allowing for robust segmentation with fewer exposures and reducing power consumption, using a fully convolutional neural network for classification and tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple exposures are used to acquire depth images, then depth imaging resolution is improved, but power consumption increases linearly
Solution Approach 1:
The patent applies partial action by acquiring fewer exposures than traditionally required (using less than 6 exposures instead of the conventional 6-9 exposures). This reduces power consumption while maintaining sufficient depth imaging capability through the use of aliased depth data for segmentation tasks, where complete de-aliasing is not necessary.
Solution Approach 2:
The patent changes the parameter of exposure count from the conventional 6-9 exposures to fewer exposures (less than 6). This parameter change directly reduces power consumption since power consumption depends linearly on the number of exposures, while still achieving functional depth imaging through aliasing tolerance in segmentation algorithms.
2Measurement precision
If multiple exposures are used to acquire depth images, then depth imaging quality is improved, but frame rate is limited
Solution Approach 1:
The patent uses partial action by acquiring fewer exposures (less than 6) instead of the full conventional sequence. This reduces the time required per frame acquisition, thereby increasing frame rate while maintaining sufficient depth imaging quality for segmentation applications through aliasing-tolerant processing.
3Measurement precision
If de-aliasing is performed on depth data, then measurement accuracy is improved, but motion artifacts are introduced and processing efficiency decreases
Solution Approach 1:
The patent extracts only the necessary depth information required for segmentation tasks without performing complete de-aliasing. By taking out only the essential depth data needed for object segmentation and tracking, the system avoids introducing motion artifacts associated with full de-aliasing while maintaining sufficient measurement accuracy for the intended application.
Solution Approach 2:
The patent uses inexpensive aliased depth data directly for segmentation without investing in expensive and complex de-aliasing processing. The aliased depth images, though lower quality, are sufficient for segmentation purposes and can be discarded after use, avoiding the computational overhead and motion artifacts of de-aliasing.
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
Enables robust segmentation and tracking with reduced power consumption and increased frame rates, maintaining accurate tracking despite motion artifacts, and allowing for more efficient use of optical ToF cameras in battery-powered devices.
Implementation Method 1
Some depth cameras operate according to the optical time-of-flight (ToF) principle, where distance to each point on a surface of a photographic subject is computed based on the length of the time interval in which light emitted by the camera travels out to the point and then back to the camera.
Data Source
AI summary
A method to identify one or more depth-image segments that correspond to a predetermined object type is enacted in a depth-imaging controller operatively coupled to an optical time-of-flight (ToF) camera; it comprises: receiving depth-image data from the optical ToF camera, the depth-image data exhibiting an aliasing uncertainty, such that a coordinate (X, Y) of the depth-image data maps to a periodic series of depth values {Zk}; and labeling, as corresponding to the object type, one or more coordinates of the depth-image data exhibiting the aliasing uncertainty.