Video Coding Rate Control Using Frame Rescaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing end-to-end video coding techniques lack sufficient operational rate-distortion points in rate-distortion models, complicating rate control in video coding optimization.
Innovation Solution
A computer-implemented method that determines coding parameters using rate-distortion models and a rescale parameter to process video frames, incorporating a video compression model like a Lagrange multiplier, and adjusts frame resolution to improve rate control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing end-to-end video coding techniques are used, then video compression is achieved, but the number of operational rate-distortion points is insufficient, complicating rate control
Solution Approach 1:
The patent applies parameter changes by introducing a rescale parameter that modifies the resolution of video frames. By changing the spatial dimension parameter (resolution), the system generates multiple operational rate-distortion points from a single rate-distortion model, enabling accurate rate control without increasing model complexity. The rescale parameter allows the same model to operate at different resolutions, effectively multiplying the available operating points.
2Measurement precision
If multiple rate-distortion models are trained to achieve denser coding bitrate points, then rate control precision improves, but model complexity and training requirements increase
Solution Approach 1:
The patent implements universality by designing a single rate-distortion model that can serve multiple bitrate points through the rescale parameter. Instead of requiring separate models for different bitrates, the same model becomes multi-functional by operating at different resolutions. This universal approach allows one model to replace what would traditionally require multiple specialized models, achieving dense bitrate control without increasing the number of models.
Solution Approach 2:
The patent introduces another dimension (spatial resolution) to the rate-distortion optimization problem. By adding the rescale parameter that controls frame resolution, the system transforms a one-dimensional bitrate optimization into a two-dimensional optimization space (bitrate × resolution). This dimensional expansion allows the generation of multiple operational points from a single model without requiring multiple models.
3Productivity
If resolution of frames is changed to facilitate encoding, then coding efficiency improves, but additional parameter management is required
Solution Approach 1:
The patent merges the resolution adjustment function with the rate-distortion optimization process. Instead of treating resolution change as a separate preprocessing step, the rescale parameter is integrated into the rate-distortion model itself. This merging allows resolution adjustment and rate optimization to occur simultaneously within a unified framework, improving coding efficiency while managing parameter complexity through integration rather than separation.
Data Source
AI summary
A computer-implemented method for processing a video includes: (a) determining, based on one or more rate-distortion models and number of bits for a frame of the video, coding parameters for processing the frame, the coding parameters comprising a rescale parameter r and a video compression model λ, and (b) processing the frame based on the rescale parameter r and the video compression model λ determined in (a) to form at least part of a bitstream of the video.