Adaptive Chroma Reshaping for HDR Image Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current chroma reshaping techniques for HDR images often result in oversaturated and desaturated pictures of poor visual quality, especially when trying to maintain backward compatibility with SDR decoders, as they fail to accurately reconstruct original HDR signals due to oversaturation and clipping issues.
Innovation Solution
The proposed solution involves adaptive chroma reshaping techniques that use a processor to generate maximal scale factors and forward reshaping functions based on luma values, partitioning pixel values into bins to determine t-percentile values and compute optimal scale factors, and constructing new forward reshaping functions that are smooth and reversible to prevent clipping, allowing for efficient encoding and decoding of HDR content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional chroma reshaping techniques are used for HDR images, then backward compatibility with SDR decoders is maintained, but the images become oversaturated and desaturated resulting in poor visual quality
Solution Approach 1:
The patent applies local quality by dividing the luma range into multiple bins and computing different scale factors for each bin based on local chroma pixel value statistics. This allows adaptive chroma reshaping that adjusts saturation levels locally across different luminance regions, preventing both oversaturation in mid-tone regions and desaturation in high-contrast regions, while maintaining compatibility with SDR decoders.
Solution Approach 2:
The patent implements dynamics by making the chroma scale factors adaptive rather than fixed. The scale factors are dynamically computed based on the distribution of chroma pixel values within each luma bin, allowing the reshaping function to adapt to different image content characteristics. This dynamic adjustment enables the system to maintain optimal image quality across diverse HDR content while preserving backward compatibility.
2Productivity
If fixed chroma scale factors are applied to maintain compatibility with SDR systems, then encoding efficiency is improved, but clipping and oversaturation occur resulting in loss of HDR signal information
Solution Approach 1:
The patent applies parameter changes by varying the chroma scale factors based on the statistical properties of chroma pixel values within each luma bin. Instead of using a fixed scale factor, the system computes optimal scale factors that adapt to the specific content characteristics, thereby preventing clipping and preserving HDR signal information while maintaining encoding efficiency through the use of reshaping functions.
Solution Approach 2:
The patent implements feedback by using the actual chroma pixel value distributions observed in the input HDR image to determine the scale factors. The system analyzes the chroma statistics within each luma bin and uses this feedback information to compute appropriate scale factors, ensuring that the reshaping process accurately represents the original HDR signal without loss of information.
3Manufacturing precision
If aggressive chroma saturation enhancement is applied to improve visual quality, then color fidelity is enhanced, but clipping occurs in high-contrast regions resulting in desaturation
Solution Approach 1:
The patent applies local quality by analyzing chroma pixel values within specific luma bins and computing localized scale factors for each bin. This allows the system to enhance color fidelity in mid-tone regions where saturation is typically desired, while avoiding aggressive enhancement in high-contrast regions where clipping would occur. The local analysis enables selective saturation enhancement that preserves chroma information across different luminance conditions.
Solution Approach 2:
The patent applies partial action by computing scale factors that provide sufficient saturation enhancement for most regions without excessively aggressive transformation that would cause clipping. The scale factors are calibrated to provide just enough enhancement to improve color fidelity while leaving headroom to prevent clipping in high-contrast regions, thereby avoiding information loss.
Data Source
AI summary
Methods and systems for adaptive chroma reshaping are discussed. Given an input image, a luma-reshaped image is first generated based on its luma component. For each chroma component of the input image, the range of the pixel values in the luma reshaped image is divided into bins, and for each bin a maximal scale factor is generated based on the chroma pixel values in the input image corresponding to the pixels of the luma reshaped image in the bin. A forward reshaping function is generated based on a reference reshaped function and the maximal scale factors, and reshaped chroma pixel values for the chroma component are generated based on the forward reshaping function and the corresponding pixel values in the luma reshaped image. Implementations options using look-up tables for mobile platforms with limited computational resources are also described.