Chroma Reshaping for HDR Video Compression Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression standards are inefficient in handling high-dynamic range (HDR) images, as they often require more bits for chroma information, leading to motion-blurred pictures and increased compression complexity, while HDR images in formats like IPT-PQ or ICTCP are compression unfriendly compared to traditional YCbCr formats.
Innovation Solution
The method involves chroma reshaping to modify the statistical characteristics of HDR video signals, converting them into codec-friendly formats like YCbCr, which reduces the energy required for encoding, thereby retaining sharpness and detail, by applying reshaping functions that match the statistical characteristics of standard dynamic range signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If HDR images are encoded using existing video compression standards, then chroma information quality is maintained, but bit usage increases and compression efficiency decreases
Solution Approach 1:
The patent applies parameter changes by transforming HDR chroma values through a reshaping function that maps them to a different statistical distribution. This transformation changes the parameters of the chroma signal to match the statistical characteristics of SDR signals, enabling more efficient compression while maintaining visual quality.
Solution Approach 2:
The patent introduces an intermediary reshaping function that acts as a mediator between HDR and SDR chroma representations. This function transforms HDR chroma values into a format that is more compatible with existing compression standards, serving as a bridge that enables efficient compression without direct loss of chroma information quality.
2Measurement precision
If HDR images are encoded using existing video compression standards, then chroma information is preserved, but compression complexity increases
Solution Approach 1:
By changing the statistical parameters of HDR chroma through the reshaping function, the patent makes the signal more compatible with existing compression standards. This parameter transformation reduces the complexity of compression operations while preserving chroma information, as the reshaped signal follows a distribution that existing codecs are optimized to handle.
3Productivity
If chroma reshaping is applied to HDR signals, then compression efficiency improves, but statistical characteristics change
Solution Approach 1:
The patent deliberately applies parameter changes through the chroma reshaping function to improve compression efficiency. The reshaping transforms HDR chroma statistical characteristics to match SDR characteristics, which are more favorable for existing compression standards. This controlled change in statistical parameters enables better compression performance while maintaining visual fidelity.
4Measurement precision
If HDR chroma values are directly encoded without reshaping, then original signal integrity is maintained, but motion blur increases
Solution Approach 1:
The patent applies parameter changes to HDR chroma values through reshaping, which improves picture sharpness by reducing motion blur artifacts. The reshaped chroma signal has statistical characteristics that are more compatible with motion compensation algorithms in video compression, resulting in sharper images during motion sequences while maintaining signal integrity through reversible transformation.
Data Source
AI summary
A first signal in a first color format is received. The first signal is transformed to a second signal in a second color format to be compressed using an encoder. Before encoding, the chroma components of the second signal are reshaped according to the statistical characteristics of the chroma components of the first signal to generate a third signal. In a decoder, an inverse reshaping functions is applied to the chroma components of the decoded signal to generate an approximation of the second signal.