API for Multiplanar Image Data Interoperability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computing systems face inefficiencies in handling and converting image and video formats, leading to waste of resources and a need for interoperability between different formats to enable seamless processing and usage.
Innovation Solution
The implementation of an API that imports and manages image data stored in multiplanar formats, such as NV12, by mapping planes of intensity and color values into separate CUDA arrays, allowing for efficient processing and interoperability between different APIs and contexts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If image or video format conversion is performed to enable computing resource usage, then interoperability is improved, but computing resources are wasted
Solution Approach 1:
The patent introduces an intermediary layer (the API and multiplanar format specification) that enables direct usage of image and video data between different computing resources and APIs without requiring format conversion. This mediator allows NV12 and other multiplanar formats to be accessed directly by CUDA and other computing resources, eliminating the need for conversion while maintaining interoperability.
Solution Approach 2:
The patent creates a universal multiplanar format specification that can be used across multiple different APIs and computing resources simultaneously. By defining a standardized way to represent image and video data in multiplanar formats, the same data structure can be directly consumed by different computing resources without conversion, making the format universally compatible while avoiding resource waste.
2Adaptability or versatility
If format conversion is performed to enable computing resource usage, then compatibility is improved, but processing time is increased
Solution Approach 1:
The patent applies preliminary action by pre-organizing image and video data into the multiplanar format (such as NV12) at the source, with clear plane separations and standardized memory layouts. This preliminary structuring allows computing resources to directly access and process the data without requiring any format conversion operations, thereby eliminating processing time delays while maintaining compatibility.
3Productivity
If full image data is loaded for processing, then processing completeness is improved, but memory usage is increased
Solution Approach 1:
The patent segments image and video data into separate planes (luminance Y plane and chrominance UV planes in NV12 format, or multiple Y planes in YV12 format). This segmentation allows computing resources to load and process only the specific planes needed for a given task, rather than requiring the entire image data in a single unified format, thereby reducing memory usage while maintaining processing completeness for the required data.
Data Source
AI summary
Apparatuses, systems, and techniques to perform an API to retrieve a portion of an image based, at least in part, on an indication of the portion of said image. In at least one embodiment, an API uses a handle that references a memory location for said image, and said image has a first plane corresponding to image data (e.g., color) and a second plane corresponding to different image data (e.g., intensity). Apparatuses, systems, and techniques to perform an API to retrieve a portion of an image based, at least in part, on an indication of the portion of the image.