Interlaced Video Motion Prediction Using Field-Specific Temporal References
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hierarchical encoding methods with spatial scalability cannot generate motion and texture predictors for high resolution interlaced sequences from low resolution interlaced sequences, as they do not account for the interlaced nature of the sequences.
Innovation Solution
A method is developed to generate motion and texture predictors for high resolution interlaced pictures by sub-sampling motion and texture data from corresponding low resolution interlaced pictures, using inter-layer ratios to align fields and handle interlaced sequences, specifically applying the Extended Spatial Scalability (ESS) method and its modified version (MESS) to ensure valid prediction and avoid invalid motion vectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional prediction methods or inter-layer prediction methods are used for hierarchical encoding with spatial scalability, then encoding efficiency is improved, but the methods cannot generate valid motion predictors for interlaced sequences
Solution Approach 1:
The patent changes the parameter of temporal reference matching by introducing field-specific temporal reference matching. Instead of using a single temporal reference for the entire picture, the method uses different temporal references for top and bottom fields, allowing valid motion predictor generation for interlaced sequences while maintaining encoding efficiency.
Solution Approach 2:
The patent segments the interlaced picture into top and bottom fields, each with its own temporal reference. This segmentation allows the motion prediction process to handle each field separately, ensuring that motion vectors are generated only from temporally valid reference fields, thus resolving the issue of invalid motion predictors in interlaced sequences.
2Adaptability or versatility
If motion data is sub-sampled from low resolution pictures to generate high resolution predictors, then scalability is improved, but the interlaced nature of sequences causes prediction errors
Solution Approach 1:
The patent changes the sub-sampling process by applying it separately to top and bottom fields with their respective temporal references. This field-specific sub-sampling approach maintains the interlaced structure during resolution scaling, preventing prediction errors that would arise from treating interlaced sequences as progressive.
Solution Approach 2:
The patent introduces temporal reference matching as an intermediary step between low resolution motion data and high resolution prediction. By verifying temporal reference compatibility before generating motion vectors, the method ensures that only valid reference fields are used, thereby maintaining prediction accuracy during the scalability process.
Data Source
AI summary
The invention relates to a method for generating for at least one block of pixels of a picture of a sequence of interlaced pictures at least one motion predictor and at least one texture predictor from motion data, respectively texture data, associated with the pictures of a sequence of low resolution interlaced pictures.