HDR Texture Compression via Luminance-Chrominance Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High dynamic range (HDR) textures require significant storage and bandwidth, limiting the number of textures available for rendering and frame rate, especially when using complicated filtering methods, due to their large size compared to low dynamic range (LDR) textures.
Innovation Solution
HDR textures are transformed into luminance and chrominance space, with adaptive selection of RGB channels for conversion, quantized to integer format, and then compressed using DirectX texture compression (DXTC) to create a data structure that can be decompressed for real-time rendering on existing LDR hardware, increasing the texture compression ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If HDR textures are stored in floating-point format to maintain real radiance values, then imaging quality and dynamic range are improved, but storage size and memory bandwidth requirements increase significantly
Solution Approach 1:
The patent transforms HDR texture data from floating-point RGB format to a compressed integer format by converting to luminance-chrominance space and applying quantization. This parameter transformation reduces the bit depth from 32-bit floating-point to 8-bit integer per channel, achieving significant storage reduction while maintaining acceptable image quality through adaptive quantization parameters.
Solution Approach 2:
The patent introduces an intermediary compression layer between the HDR source data and the final texture storage. This intermediary process includes transforming the data to luminance-chrominance space, applying adaptive quantization, and encoding with DXTC compression algorithms. The intermediary compression layer acts as a mediator that reduces data size while preserving essential HDR characteristics.
2Measurement precision
If HDR textures are used to achieve real radiance values and improved lighting rendering, then rendering quality is improved, but frame rate decreases due to increased memory bandwidth requirements
Solution Approach 1:
The patent changes the data representation parameters by converting from floating-point to compressed integer format, reducing the memory bandwidth requirement. This parameter change allows the system to maintain HDR rendering quality while achieving frame rates comparable to LDR textures, as the compressed format reduces the data transfer burden on the memory bus.
Solution Approach 2:
The patent creates a compressed copy of the HDR texture data that maintains the essential radiance information while reducing the data volume. The compression process preserves the critical HDR characteristics needed for realistic rendering while using significantly less bandwidth, effectively copying the essential information in a more efficient format.
3Quantity of substance
If texture compression is applied to reduce storage and bandwidth requirements, then storage efficiency is improved, but compression artifacts and quality loss occur
Solution Approach 1:
The patent applies adaptive quantization as a parameter change that adjusts the compression level based on local image characteristics. By varying the quantization parameters according to the importance of different pixel regions, the method achieves high compression ratios in low-frequency areas while maintaining sharpness and detail in high-frequency areas, thus reducing quality loss.
Solution Approach 2:
The patent implements local quality preservation by applying different compression strengths to different regions of the texture. Important regions with fine details maintain higher quality, while uniform regions undergo stronger compression. This local differentiation of quality treatment minimizes overall quality loss while achieving high storage efficiency.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A method for compressing a high dynamic range (HDR) texture. A first block of texels of the HDR texture in a red-green-blue (RGB) space may be transformed to a second block of texels in a luminance-chrominance space. The first block may have red values, green values and blue values. The second block may have luminance values and chrominance values. The chrominance values may be based on a sum of the red values, a sum of the green values and a sum of the blue values. The luminance values and the chrominance values may be converted to an 8-bit integer format. The luminance values may be modified to restore a local linearity property to the second block. The second block may be compressed.