HDR Video Quality Index via Perceptual Luminance Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a lack of effective methods to quantify and measure the impact of image processing operations on High Dynamic Range (HDR) video quality, particularly in evaluating distorted HDR video sequences through both subjective and objective approaches.
Innovation Solution
A method is developed to determine a visual quality index for HDR video sequences by transforming frames into a perceived luminance domain, computing similarity frames at different spatial scales and orientations, pooling error values to generate a short-term quality score, and calculating a visual quality index based on spatio-temporal segments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If traditional LDR capturing and display devices are used, then device complexity is reduced, but luminance range and contrast capability deteriorate
Solution Approach 1:
The patent transforms HDR video data from linear luminance space to perceptual luminance space using gamma correction or logarithmic transformation. This parameter change enables the representation of wide luminance ranges in a compressed format suitable for standard display devices while preserving perceptual quality differences.
Solution Approach 2:
The patent segments the HDR video processing into distinct stages: tone mapping to compress luminance range, quality assessment metric computation in perceptual space, and distortion measurement. This segmentation allows complex HDR processing to be broken down into manageable operations that can be performed with standard computational resources.
2Measurement precision
If subjective quality assessment methods are used, then measurement accuracy is improved, but time consumption and automation capability deteriorate
Solution Approach 1:
The patent introduces an intermediary computational model that simulates human visual perception through perceptual luminance transformation and distortion metric computation. This intermediary system bridges subjective quality assessment (which is accurate but manual) and objective automated assessment (which is fast but traditionally less accurate), enabling automated evaluation with perceptual relevance.
Solution Approach 2:
The patent replaces the mechanical system of human subjectives (manual viewing and rating by human observers) with an automated computational system that uses perceptual luminance models and mathematical distortion metrics. This substitution maintains measurement precision while enabling full automation and scalability.
3Illumination intensity
If HDR video sequences with high luminance values are captured, then luminance range and contrast are improved, but information loss in dark and bright areas worsens
Solution Approach 1:
The patent applies perceptual luminance transformation (gamma correction or logarithmic encoding) to change the parameter space from linear luminance to perceptually uniform space. This transformation compresses the wide HDR luminance range into a manageable representation that preserves perceptually relevant information while reducing dynamic range, preventing information loss in both dark and bright regions.
Solution Approach 2:
The patent converts the potential harm of high luminance values exceeding display capabilities into a benefit by using tone mapping and perceptual encoding. The excessive luminance range is transformed into a compressed perceptual representation that fits within display constraints while preserving visually important information, turning the limitation into an opportunity for perceptually optimized compression.
4Productivity
If image processing operations such as compression and tone mapping are applied, then productivity and processing efficiency are improved, but video quality and distortion levels worsen
Solution Approach 1:
The patent implements a feedback mechanism where the quality assessment metric computes distortion measurements in perceptual luminance space and uses this information to evaluate and compare different processing operations. This feedback loop enables optimization of compression and tone mapping parameters to minimize perceptual distortion while maintaining processing efficiency.
Solution Approach 2:
The patent replaces traditional quality assessment mechanisms (subjective viewing) with automated computational metrics operating in perceptual space. This substitution enables rapid, high-throughput quality evaluation of processed HDR video, allowing extensive processing operations to be performed and evaluated efficiently without sacrificing quality measurement accuracy.
Data Source
AI summary
A method for determining objectively a visual quality index of at least one high dynamic range video sequence, referred to as an HDR sequence, distorted by image processing operations and issued from a reference high dynamic range video sequence, referred to a reference sequence or a reference HDR sequence. The method is based on signal pre-processing, transformation, and subsequent frequency based decomposition. Video quality is then computed based on a spatio-temporal analysis that relates to human eye fixation behavior during video viewing. One advantage of this method is that it does not involve expensive computations.