Depth Sensor Tilt Mechanism for Extended Range Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience measurement systems face inaccuracies in recognizing people and gestures due to limited depth range of depth sensors, leading to high rates of false positives, false negatives, and inaccurate identity recognitions, which affect media exposure data and ratings calculations.
Innovation Solution
The system captures multiple depth values at different angles using a tilt motor to calculate an accumulated depth value, providing a depth gradient that enhances recognition accuracy by increasing resolution for objects outside the optimal depth range, thereby improving person detection, identity determination, and gesture detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a depth sensor is used to detect objects in the environment, then person detection capability is provided, but the depth range is limited causing high rates of false positives and false negatives
Solution Approach 1:
The patent introduces angular position as an additional dimension to depth measurement. By capturing depth values at multiple angular positions (0 degrees, 30 degrees, 60 degrees) and calculating accumulated depth values, the system extends the effective depth range beyond what a single fixed-depth sensor can provide, thereby reducing false positives and false negatives in person detection.
Solution Approach 2:
The patent uses an intermediary computational process (accumulated depth value calculation) to bridge the gap between limited sensor depth range and the need for broader detection capability. The accumulated depth value serves as a mediator that combines information from multiple angular perspectives to accurately determine objects at distances beyond the optimal depth range of any single sensor position.
2Measurement precision
If multiple depth values are captured at different angles to extend depth range, then measurement precision is improved, but device complexity increases due to tilt motor and multiple measurements
Solution Approach 1:
The patent implements a dynamic tilt mechanism that adjusts the angular position of the depth sensor based on detected conditions. The system transitions from a static single-position sensor to a dynamic multi-position sensor, enabling the device to adapt its measurement angles to capture depth gradients and extend its effective measurement range while maintaining manageable complexity through conditional activation.
3Reliability
If accumulated depth values are calculated from multiple angular positions, then recognition accuracy is improved for objects outside optimal depth range, but loss of time occurs due to multiple capture cycles
Solution Approach 1:
The patent performs preliminary depth value captures at multiple angular positions before final object recognition decisions are made. By pre-capturing depth values at 0 degrees, 30 degrees, and 60 degrees and calculating accumulated depth values in advance, the system prepares enhanced depth information that can be quickly referenced during recognition, reducing the time penalty of multiple measurements.
Solution Approach 2:
The system uses feedback from the accumulated depth value calculations to determine when additional angular measurements are necessary. By analyzing the depth gradient information, the system can identify cases where objects fall outside the optimal depth range and trigger targeted additional measurements, rather than uniformly increasing measurement frequency for all scenarios, thus reducing overall time loss.
Data Source
AI summary
Methods, apparatus, and articles of manufacture to monitor environments are disclosed. An example method includes analyzing a first depth value corresponding to a coordinate of an object of an image captured by a depth sensor, the first depth value having been captured with the depth sensor positioned at a first angle relative to a reference axis; moving the depth sensor to a second angle relative to the reference axis, the second angle being different than the first angle; triggering capture of a second depth value at the coordinate with the depth sensor positioned at the second angle; and calculating a depth gradient for the coordinate based on the first and second depth values.