Intra Skip Mode for Depth Image Encoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high data requirements for storing, transmitting, and reproducing three-dimensional (3D) images, which consist of texture and depth images from multiple viewpoints, necessitate the development of efficient encoding methods to reduce data volume.
Innovation Solution
The implementation of an intra skip mode in multi-view video encoding and decoding methods, where an intra skip flag and intra skip prediction mode information are used to determine predicted values for blocks in depth images, allowing for the reconstruction of blocks without residual data in the bitstream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional video encoding methods are used for depth images, then coding efficiency is maintained at acceptable levels, but data volume remains large
Solution Approach 1:
The patent changes the encoding parameters for depth images by applying intra skip mode, which modifies how prediction is performed. Instead of using conventional prediction methods that maintain acceptable coding efficiency, the patent uses skip mode where the current block is predicted from reference blocks without calculating residual data, fundamentally changing the encoding parameter set and achieving higher compression ratios for depth images.
Solution Approach 2:
The patent extracts and removes residual data from the bitstream when intra skip mode is applied. By identifying that depth images can be adequately reconstructed using prediction alone without residual correction data, the patent extracts only the necessary prediction information and discards the residual data component, significantly reducing the overall bitstream volume.
2Productivity
If encoding methods with high distortion rates are used for depth images, then coding efficiency increases, but image quality deteriorates
Solution Approach 1:
The patent applies different encoding strategies to different image types within the same system. For depth images specifically, intra skip mode is applied which allows higher distortion, while texture images can use conventional encoding methods. This local differentiation of quality requirements based on image type enables overall system efficiency improvement without compromising where quality matters most.
Solution Approach 2:
The patent inverts the conventional approach by prioritizing coding efficiency over image quality for depth images. Instead of maintaining quality as the primary constraint and efficiency as secondary, the patent makes efficiency the primary goal for depth images where quality imperfections are less perceptible, effectively inverting the traditional quality- efficiency trade-off hierarchy.
3Manufacturing precision
If conventional encoding methods are used, then image quality is maintained, but data transmission and storage requirements increase
Solution Approach 1:
The patent uses copying by predicting the current block from reference blocks in the same or previous pictures. Instead of encoding and transmitting the actual pixel data or residual data, the patent transmits only the prediction mode information and parameters, allowing the decoder to copy the predicted values directly, significantly reducing data volume while maintaining acceptable reconstruction quality.
Solution Approach 2:
The patent performs prediction in advance during the encoding process, generating predicted block values before actual encoding. By pre-calculating what the block values should be based on reference blocks and prediction modes, the patent eliminates the need to transmit correction data, performing the useful action of data generation beforehand and only transmitting minimal control information.
Data Source
AI summary
Provided is a method of decoding a multi-view image, the method including obtaining, from a bitstream, an intra skip flag indicating whether a current block included in a depth image of the multi-view image is to be reconstructed based on an intra skip mode; obtaining, from the bitstream, intra skip prediction mode information indicating an intra prediction mode to be used for the current block, from among a plurality of intra prediction modes, when the intra skip flag indicates that the current block is to be reconstructed based on the intra skip mode; determining predicted values of samples included in the current block according to an intra prediction method indicated by the intra skip prediction mode information; and reconstructing the current block by determining reconstructed values of the samples, based on the predicted values of the samples.