Predicted Picture Generation Using Reference View Depth

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing methods for generating predicted pictures in video encoding/decoding processes rely heavily on depth information from both the reference and target views, leading to a significant data dependence and inefficiencies in encoding and decoding.

Innovation Solution

A method where the reference rectangular block of pixels is determined using the location of the target rectangular block and depth information of the reference view, without needing the depth picture of the target view, allowing forward projection to be performed using only the depth information of the reference view.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If depth information from both reference view and target view is used to determine reference block and perform forward projection, then prediction accuracy is improved, but data dependence increases and encoding/decoding efficiency deteriorates

Engineering Contradiction:
Improveprediction accuracyVSAvoidencoding and decoding efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent extracts and removes the depth information requirement from the target view, keeping only the depth information from the reference view. This extraction eliminates the data dependence on target view depth while maintaining the forward projection functionality, thus resolving the contradiction between prediction accuracy and encoding efficiency.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the depth information usage into two independent parts: depth information from reference view (retained) and depth information from target view (removed). This segmentation allows the system to maintain necessary prediction functionality while eliminating redundant data requirements, improving encoding and decoding efficiency.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If depth information from both views is simultaneously used in predicted picture generation, then prediction quality is improved, but device complexity increases

Engineering Contradiction:
Improveprediction qualityVSAvoiddata dependence
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the unnecessary component (target view depth information) from the system, reducing data dependence and simplifying the processing architecture while retaining sufficient prediction quality through reference view depth information alone.

Inventive Principle:
Principle #2Taking out (Extraction)

3Manufacturing precision

If reference block is determined using depth information from target view, then projection accuracy is improved, but encoding complexity increases

Engineering Contradiction:
Improveprojection accuracyVSAvoidencoding complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent removes the requirement for target view depth information in reference block determination, simplifying the encoding process. The reference block is determined using only reference view depth information and parameter information, reducing encoding complexity while maintaining adequate projection accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2903283B1Prediction image generation method and device
Publication Date: 2021.11.03 ZTE CORP
  • EP2903283B1 patent drawingFigure 1~2
  • EP2903283B1 patent drawingFigure 3
  • EP2903283B1 patent drawingFigure 4~5

AI summary

Disclosed are a method and device for generating a predicted picture, the method comprising: determining a reference rectangular block of pixels according to parameter information which includes a location of a target rectangular block of pixels and/or depth information of a reference view; mapping the reference rectangular block of pixels to a target view according to the depth information of the reference view to obtain a projection rectangular block of pixels; and acquiring a predicted picture block from the projection rectangular block of pixels. The technical problem of relatively large dependence among the data brought by simultaneously employing the depth picture of the target view and the depth picture of the reference view in the process of generating the predicted picture in the prior art is solved, and the technical effects of reducing the dependence on the data and improving the encoding and decoding efficiency is achieved.