HDR Tone Mapping via Luminance Layer Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional tone mapping techniques for High Dynamic Range (HDR) images often clip details due to the use of a single, linear scaling factor, which limits the representation of contrast variations and results in a loss of image quality when converting HDR images to lower dynamic range formats.
Innovation Solution
A tone mapping workflow that separates HDR images into luminance and color components, processes the luminance to create a base layer and detail layer, and applies a nonlinear remapping function to compress the base layer, leaving space at the top of the dynamic range for the detail layer, allowing for more accurate representation of image details.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single linear scaling factor is used to compress the base layer, then the dynamic range is reduced to fit the output medium, but image details are clipped and lost
Solution Approach 1:
The luminance component is segmented into base layer and detail layer through high-pass filtering. The base layer contains low-frequency information and is compressed using a nonlinear remapping function, while the detail layer contains high-frequency information and is preserved separately. This segmentation allows different processing strategies for different frequency components, preventing detail loss during compression.
Solution Approach 2:
A nonlinear remapping function is applied to the base layer instead of a linear scaling factor. The remapping function uses parameters such as gamma values and threshold settings to control the compression characteristics, allowing preservation of highlight details while still achieving dynamic range reduction. The detail layer is then blended back with the compressed base layer to restore fine details.
2Adaptability or versatility
If the base layer is compressed to the full output range, then the dynamic range is maximized, but space is lost for the detail layer
Solution Approach 1:
The image is segmented into base layer and detail layer, allowing independent processing. The base layer is compressed to occupy the lower portion of the output dynamic range, while the detail layer is preserved in the upper portion. This segmentation enables both full dynamic range utilization and detail preservation simultaneously.
Solution Approach 2:
The solution moves from a single-dimensional compression approach to a two-dimensional approach by separating the luminance into base and detail layers. This dimensional separation allows the base layer to be compressed nonlinearly while leaving headroom in the dynamic range for the detail layer, effectively adding a layer dimension to the compression process.
3Productivity
If conventional tone mapping is used, then processing is simple and fast, but creative control and output flexibility are limited
Solution Approach 1:
By segmenting the luminance into base and detail layers, the patent enables independent processing of each layer. Users can apply different adjustments to the base layer (exposure, contrast, tone curve) and detail layer (sharpness, detail enhancement) separately, providing creative control while maintaining efficient processing through the established layer separation framework.
Solution Approach 2:
The patent introduces dynamic, adjustable parameters for the tone mapping process including gamma values, threshold settings, and blending factors. These parameters can be modified to achieve different creative effects while maintaining the efficient base-detail layer processing structure, thus balancing productivity with adaptability.
Data Source
AI summary
Methods, apparatus, and computer-readable storage media for tone mapping High Dynamic Range (HDR) images. The HDR image is separated into luminance and color. Luminance is processed according to the parameters to obtain a base layer and a detail layer. The base layer is compressed into a lower dynamic range and the detail layer is adjusted according to the parameters. The compressed base layer, the detail layer, and the color component may be output as separate layers, and various image processing tools and techniques may be applied to the component layers separately to modify the layer(s). One or more tone-mapped images may be generated by merging the modified layers. Thus, each layer of the tone-mapped image may be processed separately using various image processing tools or techniques to modify the output of the tone mapping technique in a wide variety of ways.