Image Codec Rate Control Using Fixed-Point Parameter Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image and video encoding technologies, such as hybrid codecs and neural network-based approaches, face challenges in optimizing rate control for bitrate and quality, leading to inefficiencies in signaling and computational complexity.
Innovation Solution
A method for encoding and decoding images using fixed-point representations of rate control parameters, specifically the displacement between target and reference rate control parameters, reducing the number of bits required for signaling and eliminating the need for floating-point calculations, thereby enhancing computational efficiency and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If floating-point rate control parameters are used in neural network-based image encoding, then rate control precision is improved, but computational complexity and device compatibility deteriorate
Solution Approach 1:
The patent changes the parameter representation from floating-point to fixed-point format. Specifically, rate control parameters are represented as fixed-point numbers with a predetermined number of integer bits and fractional bits, allowing precise rate control while maintaining computational simplicity and device compatibility throughout the encoding process
2Measurement precision
If precise rate control parameters are signaled in the bitstream, then rate control accuracy is improved, but signaling overhead increases
Solution Approach 1:
The patent changes the parameter format from floating-point to fixed-point representation, which can be more efficiently encoded. The fixed-point parameters require fewer bits to represent the same precision level, reducing signaling overhead in the bitstream while maintaining rate control accuracy
Solution Approach 2:
The patent extracts only the essential rate control information needed for decoding by using fixed-point representation, which separates the integer and fractional parts clearly. This allows the encoder to signal only the necessary fixed-point values without transmitting redundant floating-point data structures
3Measurement precision
If floating-point arithmetic is used for rate control calculations, then calculation precision is improved, but processing speed and stability deteriorate
Solution Approach 1:
The patent changes the arithmetic parameter from floating-point to fixed-point format. Fixed-point arithmetic operations are inherently faster and more stable than floating-point operations because they use simple integer arithmetic with predetermined decimal positions, eliminating the need for complex floating-point hardware and reducing processing time while maintaining sufficient precision for rate control calculations
Data Source
AI summary
The present disclosure provides techniques for improving signaling efficiency in the context of rate control governing a desired ratio of bitrate load and quality in image compression. It is provided a method for encoding an image comprising obtaining the image, encoding the image into a bitstream based on a first coding parameter, wherein the first coding parameter is indicative of a first displacement between a first target rate control parameter and a first reference rate control parameter, and encoding the first coding parameter into the bitstream. Further, it is provided a method for decoding an image, comprising receiving a bitstream comprising coded data of the image, parsing the bitstream to obtain a first coding parameter, wherein the first coding parameter is indicative of a first displacement between a first target rate control parameter and a first reference rate control parameter, and reconstructing the image based on the first coding parameter.