Spatial Audio Positioning Using Last-Known Direction Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In augmented reality, managing spatial audio presentation is challenging when the location of real-world objects is outside the limited field of view of sensors, leading to inaccurate audio positioning and potential user confusion.
Innovation Solution
An apparatus with a processor and memory configured to identify and track movable real-world objects using sensors with a limited field of view, employing a last-known-direction for spatial audio placement until the object is re-identified, and adjusting the audio direction based on user head orientation and sensor data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If sensors with limited field of view are used to track real-world objects, then device complexity is reduced, but spatial audio positioning accuracy deteriorates when objects move out of the sensor's field of view
Solution Approach 1:
The system performs preliminary tracking of real-world objects using sensors within the limited field of view, storing the last known position and direction. When an object moves out of the field of view, the spatial audio presentation continues using this pre-stored position information, maintaining audio continuity without requiring complex omnidirectional sensor systems.
Solution Approach 2:
The patent introduces an intermediary computational process that bridges the gap between limited sensor data and continuous spatial audio presentation. This intermediary system uses the last known direction and position information to maintain accurate spatial audio placement even when objects are not currently visible to the sensors, resolving the contradiction between simple sensing and accurate positioning.
2Duration of action of stationary object
If the spatial audio presentation continues using last-known-direction when objects are not visible, then audio continuity is maintained, but potential user confusion arises from audio not matching actual object location
Solution Approach 1:
The system dynamically adjusts the spatial audio presentation based on the visibility status of tracked objects. When objects are within the sensor field of view, audio positioning is updated in real-time. When objects move out of view, the system transitions to using last-known-direction information, creating a dynamic adaptation that maintains continuity while managing the loss of location information.
Solution Approach 2:
The patent applies beforehand cushioning by maintaining spatial audio presentation using last-known-direction information as a buffer when objects move out of the sensor's field of view. This cushioning approach prevents complete audio interruption and maintains immersive experience, while the system continuously searches for object re-identification to update positioning accuracy.
3Measurement precision
If multiple sensors are deployed to cover the entire scene, then object tracking accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent makes the limited-field-of-view sensors universal by combining them with computational methods that extend tracking capability beyond the physical sensor boundaries. The same sensors serve dual purposes: direct object tracking within the field of view and providing reference data for inferential tracking of objects outside the field of view through last-known-position calculations.
Solution Approach 2:
The system creates a computational copy of the spatial environment using data from the limited sensor field of view. This virtual copy includes position and direction information that is used to reconstruct spatial audio presentations for objects outside the physical sensor range, effectively copying the tracking functionality without requiring physical sensors in all directions.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
In respect of spatial audio configured to be associated with a moveable, particular object in a scene, the spatial audio for presentation such as to be perceived as originating from a particular direction, the location of the object determined based on automatic identification in sensor data to enable positioning of the spatial audio to correspond to the location of the object, the sensor data from a sensor having a limited field of view of the real-world scene at any one time; based on a current location of the object being unknown; provide for audible presentation of the spatial audio with a previously-determined-direction and, based on a current field of view of the user moving to the previously-determined-direction, provide for modification of the spatial audio from the previously-determined-direction to a direction outside the current field of view of the user at least until the object is once again identified.