Threaded Video Encoding with Tile-Based Control Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional frame-based video processing techniques require significant data storage and transmission resources due to the large volume of data needed for each pixel in each frame, leading to memory and network burdens and latency issues.
Innovation Solution
Represent digital video as 'threads' using control points and mathematical constructs like cubic curves and splines, encoding video data on a tile basis rather than a pixel-by-pixel basis, allowing for sparser data storage and transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frame-based video encoding is used to ensure complete pixel data storage, then video fidelity is maintained, but data storage volume and transmission bandwidth requirements increase significantly
Solution Approach 1:
The patent extracts only the essential visual information from complete frame data by identifying and storing only changed regions (tiles) and their corresponding control points. Instead of storing all pixel data, it extracts minimal necessary data (control points for changed tiles) to represent video content, achieving compression while maintaining perceptual fidelity.
Solution Approach 2:
The patent segments video frames into multiple tiles and processes each tile independently. By dividing the frame into smaller regions and only encoding changed tiles with control points rather than complete pixel data, it reduces overall data volume while preserving visual quality in changed regions.
2Measurement precision
If all pixel data for each frame is stored to maintain video quality, then video fidelity is preserved, but memory storage space is consumed
Solution Approach 1:
The patent extracts only the essential visual information from complete frame data by identifying and storing only changed regions (tiles) and their corresponding control points. Instead of storing all pixel data, it extracts minimal necessary data (control points for changed tiles) to represent video content, achieving compression while maintaining perceptual fidelity.
Solution Approach 2:
The patent segments video frames into multiple tiles and processes each tile independently. By dividing the frame into smaller regions and only encoding changed tiles with control points rather than complete pixel data, it reduces overall data volume while preserving visual quality in changed regions.
3Measurement precision
If complete frame data is transmitted over the network to ensure video quality, then video fidelity is maintained, but network latency and transmission losses increase
Solution Approach 1:
The patent extracts only the essential visual information from complete frame data by identifying and storing only changed regions (tiles) and their corresponding control points. Instead of storing all pixel data, it extracts minimal necessary data (control points for changed tiles) to represent video content, achieving compression while maintaining perceptual fidelity.
Solution Approach 2:
The patent segments video frames into multiple tiles and processes each tile independently. By dividing the frame into smaller regions and only encoding changed tiles with control points rather than complete pixel data, it reduces overall data volume while preserving visual quality in changed regions.
4Loss of information
If frame-based video encoding is used to capture all pixel information, then complete video data is stored, but data processing and storage burdens on hardware increase
Solution Approach 1:
The patent extracts only the essential visual information from complete frame data by identifying and storing only changed regions (tiles) and their corresponding control points. Instead of storing all pixel data, it extracts minimal necessary data (control points for changed tiles) to represent video content, achieving compression while maintaining perceptual fidelity.
Solution Approach 2:
The patent segments video frames into multiple tiles and processes each tile independently. By dividing the frame into smaller regions and only encoding changed tiles with control points rather than complete pixel data, it reduces overall data volume while preserving visual quality in changed regions.
Data Source
AI summary
In some aspects, a method is described. The method can include receiving frame-based video comprising a series of frames, encoding the frame-based video into threaded video, and storing the thread based data in a memory device, where the threaded video is sparser than the frame-based video. Encoding the frame-based video into threaded video can include converting the frame-based video into a domain space, and generating a set of tiles, each tile of the set of tiles corresponding to a sub-space within the series of frames. The method can include generating a set of control points while traversing a driving parameter of the series of frames based on a delta.