Virtual Temporal Affine Candidates for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression systems, such as those using the HEVC and VTM standards, face limitations in creating a diverse set of affine motion candidates for improved motion compensation, leading to inefficiencies in encoding and decoding processes.
Innovation Solution
The method involves determining and selecting control point motion vectors for affine motion compensation, creating virtual temporal and spatial affine candidates, and filtering these candidates to maintain a relevant subset for encoding and decoding, thereby enhancing the affine motion model list.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If a diverse set of affine motion candidates is created to improve motion compensation, then compression efficiency is improved, but device complexity increases
Solution Approach 1:
The candidate list is segmented into virtual temporal candidates and virtual spatial candidates, with each type generated using specific rules. This segmentation allows the system to manage complexity by organizing candidates into distinct categories with dedicated generation methods, while still providing diversity for improved motion compensation.
Solution Approach 2:
Virtual affine candidates are constructed in advance using predetermined rules based on motion vectors from reference blocks, before actual motion compensation is needed. This preliminary construction reduces runtime complexity while ensuring diverse candidate availability for selecting the best motion compensation strategy.
2Measurement precision
If the number of affine motion candidates is increased to improve motion model accuracy, then encoding accuracy is improved, but processing time increases
Solution Approach 1:
Virtual affine candidates are created by copying and adapting motion vectors from previously decoded reference blocks rather than performing new motion estimation. This copying approach maintains accuracy by reusing proven motion information while significantly reducing processing time compared to generating entirely new candidates.
Solution Approach 2:
The system changes parameters of existing motion vectors (such as adjusting control point positions and motion vector values) to generate diverse virtual candidates. This parameter manipulation allows rapid generation of multiple candidates with varying characteristics, improving accuracy without the computational cost of full motion estimation for each candidate.
3Reliability
If virtual temporal and spatial affine candidates are created to enhance candidate diversity, then motion compensation quality is improved, but algorithm complexity increases
Solution Approach 1:
The system dynamically selects between virtual temporal candidates and virtual spatial candidates based on the specific coding context and block characteristics. This dynamic selection adapts the candidate generation process to different scenarios, improving motion compensation quality while managing algorithm complexity by activating only the most appropriate candidate types for each situation.
Solution Approach 2:
Virtual affine candidates serve as intermediaries between reference block motion information and the final motion compensation result. These intermediate candidates provide a bridge that transforms raw motion vector data into diverse, refined options, improving the quality of motion compensation while keeping the overall algorithm manageable through structured intermediate processing steps.
Data Source
AI summary
Methods and apparatus for creating additional affine candidates. Virtual and temporal candidates are determined using neighboring spatial and temporal sub-blocks. The sub-blocks are examined in an order known to both an encoder and a decoder. Valid sub-blocks are used to compute an affine model. The candidates can be filtered and added to a candidate list conditionally based on various criteria. The candidates can be used to determine control point motion vectors and a motion flow field can be determined. Motion vectors for sub-blocks within a video coding block can be determined. Motion compensation can be performed using the improved affine candidates and encoding/decoding based on the improved affine motion compensation.