Video Block Prediction With IBC-LIC for Illuminance Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital video compression technologies face challenges in efficiently handling illuminance changes and color differences in video content, leading to increased residual information and reduced encoding efficiency, particularly in scenarios with similar textures and varying illuminance.
Innovation Solution
The proposed solution involves a decoding method that utilizes intra block copy local illuminance compensation (IBC-LIC) to determine a template region for establishing a model, predicting a block based on IBC mode, and performing illuminance compensation to improve decoding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing digital video compression standards are used, then video data can be compressed for transmission and storage, but decoding performance is insufficient and residual information increases in scenarios with illuminance changes
Solution Approach 1:
The patent applies illuminance compensation by calculating scaling factors and offset values to adjust the luminance values in the predicted block. This transforms the parameter values of the pixel data to compensate for illuminance changes, thereby reducing residual information and improving decoding performance in scenarios with varying light conditions
Solution Approach 2:
The patent introduces an illuminance compensation mechanism as an intermediary step between prediction and reconstruction. By calculating compensation values based on template regions and applying them to the predicted block, it mediates the discrepancy between the predicted and actual blocks, reducing residual information
2Productivity
If illuminance compensation is performed on predicted blocks, then encoding efficiency improves and residual information reduces, but decoding time increases
Solution Approach 1:
The patent performs illuminance compensation during the encoding phase to generate compensation values and apply them to the predicted blocks. By completing the compensation action in advance during encoding, the decoding process can directly use the compensated values without performing additional calculations, thereby improving encoding efficiency while minimizing the impact on decoding time
Data Source
AI summary
The embodiments of the present application belong to the technical field of encoding and decoding. Provided are a decoding method, an encoding method, decoders and encoders. The decoding method comprises: determining whether a current block uses an IBC-LIC mode; if it is determined that the current block uses the IBC-LIC mode, determining at least one template area used when performing illumination compensation on the current block, the at least one template area being used for determining a first model; on the basis of the IBC mode used by the current block, performing prediction on the current block to obtain a first prediction block of the current block; and, on the basis of the first model, performing illumination compensation on the first prediction block to obtain a second prediction block of the current block.