Luma Look Matching via Precomputed Reshaping Lookup Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in encoding and decoding reversible production-quality reshaped video data to support a wide variety of display devices, particularly in transitioning between high dynamic range (HDR) and standard dynamic range (SDR) formats, which limits compatibility and quality in video content delivery.
Innovation Solution
The method involves generating multiple luma histograms based on high dynamic range images and candidate parameter values, comparing them with reference tone-mapped images to select optimal parameter values for reshaping, and using these values to create forward and backward reshaping functions for efficient encoding and decoding of video data, ensuring compatibility across different dynamic ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional encoding methods are used for video data, then compatibility with standard displays is maintained, but quality and artistic intent are lost during dynamic range transitions
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing forward reshaping functions and backward reshaping functions in lookup tables during the encoding phase. These pre-computed functions enable efficient real-time conversion between HDR and SDR formats without losing quality or artistic intent, resolving the contradiction between maintaining high video quality and ensuring broad display compatibility.
2Manufacturing precision
If histogram-based luma look matching is used to optimize reshaping parameters, then video quality and artistic intent are preserved, but computational complexity increases
Solution Approach 1:
The patent reduces computational complexity by applying preliminary action - pre-computing histogram-based luma look matching results and storing optimal reshaping parameters in lookup tables during encoding. This allows the decoding phase to simply retrieve pre-determined parameters rather than performing complex histogram matching in real-time, thus maintaining high luma look matching accuracy while significantly reducing device complexity.
Solution Approach 2:
The patent uses copying by creating lookup tables that store pre-computed reshaping functions and parameters. Instead of recalculating complex histogram-based optimizations for every video frame, the system copies and applies pre-determined transformation parameters from the lookup tables, maintaining accuracy while reducing computational burden.
3Manufacturing precision
If reversible reshaping functions are implemented, then artistic intent and video quality are preserved across dynamic ranges, but encoding and decoding complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing both forward reshaping functions (for HDR to SDR conversion) and backward reshaping functions (for SDR to HDR reconstruction) during the encoding phase and storing them in lookup tables. This allows the decoding process to simply retrieve and apply pre-computed parameters, maintaining artistic intent preservation while significantly reducing the complexity of real-time encoding and decoding operations.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
In a method to reconstruct a high dynamic range video signal, a decoder receives parameters in the input bitstream to generate a prediction function. Using the prediction function, it generates a first set of nodes for a first prediction lookup table, wherein each node is characterized by an input node value and an output node value. Then, it modifies the output node values of one or more of the first set of nodes to generate a second set of nodes for a second prediction lookup table, and generates output prediction values using the second lookup table. Low-complexity methods to modify the output node value of a current node in the first set of nodes based on computing modified slopes between the current node and nodes surrounds the current node are presented.