Inverse Tone Mapping With Histogram-Guided Highlight Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing inverse tone mapping methods struggle to effectively control the appearance of HDR images generated from LDR or SDR images, often resulting in bright areas that exceed display capabilities or dazzle viewers, and lack adaptability to varying content characteristics.
Innovation Solution
A method that adapts the gain function to respect diffuse white constraints by identifying and modifying the gain curve to ensure that the maximum luminance of HDR images does not exceed predefined display limits and adjusts for large bright areas, using histogram analysis to identify and aggregate luminance contributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If global luminance expansion is applied to extend dynamic range, then the luminance range of colors is expanded, but bright areas may exceed display capabilities and dazzle viewers
Solution Approach 1:
The patent applies different expansion strategies to different regions of the image. A luminance expansion map is generated where each pixel is associated with an expansion value, allowing local adaptation of the expansion curve. Bright regions use steeper expansion curves while darker regions use gentler curves, preventing excessive brightness in highlights while maintaining dynamic range extension throughout the image.
Solution Approach 2:
The patent dynamically adjusts the expansion curve based on local luminance characteristics. The expansion values are not fixed but are computed adaptively for each pixel or region, allowing the system to respond to varying content characteristics in real-time and optimize the balance between dynamic range extension and brightness control.
2Device complexity
If simple clipped region detection and expansion is used, then the processing complexity is reduced, but sufficient control over image appearance is not achieved
Solution Approach 1:
The patent segments the image processing into multiple controllable stages: generating a luminance expansion map, detecting clipped regions, and applying region-specific expansion curves. This segmentation allows independent control of different processing parameters and enables fine-tuned adjustment of image appearance while maintaining manageable complexity through modular processing steps.
Solution Approach 2:
The patent changes multiple parameters including expansion values, curve steepness, and region thresholds to achieve precise control over image appearance. By adjusting these parameters adaptively based on content characteristics, the system provides sophisticated control capabilities without requiring overly complex processing architecture.
3Illumination intensity
If existing ITMO methods are used, then the dynamic range is expanded, but the methods rely on strict assumptions about content and are not adaptable to varying content characteristics
Solution Approach 1:
The patent implements a self-adaptive system where the luminance expansion map is generated automatically from the input image itself, and the expansion parameters are determined by analyzing the image's own luminance distribution and clipped region characteristics. This eliminates the need for strict external assumptions about content type or characteristics, allowing the method to adapt automatically to varying content while maintaining extended dynamic range.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A method for inverse tone mapping comprising: obtaining (41) an histogram of a low dynamic range image, called LDR image; obtaining (42) an ITMO function, allowing obtaining a pixel value of a High Dynamic Range image, called HDR image, from a pixel value of the LDR image and a gain function depending on said pixel value of the LDR image; applying a search process using the obtained histogram to identify areas of the LDR image producing bright areas in the HDR image when the ITMO function is applied on said LDR image, using information representative of the bright areas to determine when modifying the gain function to insure the HDR image respects at least one predefined light energy constraint.