Audio Object Position Encoding for Low-Bandwidth Free-Viewpoint Playback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The processing load and information transfer volume on the content distribution side increase significantly with the number of listeners in free viewpoint audio reproduction, especially when multiple listeners are involved, due to the need for frequent updates of audio object positions using MPEG-H rendering.
Innovation Solution
The system adjusts the quantized accuracy of position information based on the distance between the listener and the object, transmitting low accuracy object positions initially and additional difference information as needed, allowing clients to calculate accurate polar coordinates, thereby reducing processing load and information transfer volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If position information of audio objects is updated and transferred every time positions change in free viewpoint audio reproduction, then accurate audio reproduction is achieved, but processing load and information transfer volume increase significantly
Solution Approach 1:
The position information is segmented into two parts: a base position (e.g., integer coordinates) and an offset (e.g., fractional coordinates). The base position is transferred with lower accuracy, while the offset provides the additional precision needed. This segmentation allows the system to maintain accurate audio reproduction by combining both parts on the client side, while significantly reducing the total information transfer volume compared to transferring complete high-precision positions continuously.
Solution Approach 2:
Different levels of position accuracy are applied based on local needs. Objects that are stationary or far from the listener use lower precision position information, while objects that are moving or close to the listener use higher precision (base position + offset). This local quality approach ensures accurate audio reproduction only where necessary, reducing overall processing load and information transfer volume.
2Measurement precision
If position information of audio objects is updated and transferred every time positions change in free viewpoint audio reproduction, then accurate audio reproduction is achieved, but processing load on distribution side increases
Solution Approach 1:
The position update process is segmented into base position updates and offset updates. The server performs lighter processing by calculating and transferring only the offset (difference from base position) rather than complete high-precision positions. This reduces the computational burden on the distribution side while still enabling accurate audio reproduction when base position and offset are combined on the client side.
Solution Approach 2:
The base position is calculated and established in advance as a reference frame. Subsequent position updates only require calculating the offset from this pre-established base position, rather than recalculating complete positions. This preliminary action reduces the processing load on the distribution side for each position update event.
3Quantity of substance
If low accuracy quantized position information is used, then information transfer volume is reduced, but position precision deteriorates
Solution Approach 1:
Position information is divided into a base position (transferred with lower accuracy to reduce information volume) and an offset (transferred separately to restore precision). The base position provides the coarse location, while the offset provides the fine adjustment needed for accurate audio reproduction. This segmentation allows the system to achieve both reduced information transfer volume and maintained position precision.
Solution Approach 2:
The system changes the representation parameter of position information from a single high-precision value to a composite structure of base position plus offset. This parameter change allows the same position to be represented with fewer bits for the base component, while the offset component compensates for the reduced precision, achieving both compression and accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present technology relates to an information processing apparatus, an information processing method, and a program for achieving reduction of a processing load on a distribution side along with reduction of a transfer volume of information. The information processing apparatus includes an acquisition unit that acquires low accuracy position information having first accuracy and indicating a position of an object within a space where a user is located and acquires additional information for obtaining position information that has second accuracy higher than the first accuracy, indicates the position of the object within the space, and corresponds to a position of the user and a position information calculation unit that obtains the position information on the basis of the low accuracy position information and the additional information. The present technology is applicable to an information processing apparatus.