Reference Picture List Signaling for Slice-Type Collocated Picture Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video coding techniques in VVC Draft 8 have inefficiencies and potential problems in the signaling of reference picture lists, particularly in the derivation of collocated pictures for temporal motion vector prediction, which can restrict the encoding process and introduce artifacts.
Innovation Solution
A video decoder is configured to infer the value for the syntax element indicating the reference picture list for temporal motion vector prediction based on the slice type, overriding the PH-level syntax for P slices, allowing for more flexible encoding and decoding without undue restrictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the PH-level syntax element is used to indicate the reference picture list for temporal motion vector prediction, then the encoding flexibility is improved, but the decoding complexity and potential artifacts increase due to restrictive signaling
Solution Approach 1:
The patent extracts the reference picture list indication from the PH-level syntax element and replaces it with a default value derived from slice type. This removes the problematic signaling mechanism while maintaining the essential functionality through inference-based defaults, thereby reducing decoding complexity without sacrificing encoding flexibility.
Solution Approach 2:
The system enables the decoder to self-determine the reference picture list by inferring from the slice type (P slice or B slice) without requiring explicit PH-level signaling. This self-service mechanism eliminates the need for complex syntax parsing while maintaining adaptability through context-based defaults.
2Measurement precision
If explicit signaling of collocated picture information is performed in the picture header, then the precision of temporal motion vector prediction is improved, but the bitstream overhead and encoding complexity increase
Solution Approach 1:
The patent applies local quality by differentiating the treatment of P slices and B slices through slice type inference. Instead of uniform explicit signaling, the system tailors the reference picture list selection to the specific slice type, achieving precise temporal motion vector prediction where needed while avoiding unnecessary overhead in other cases.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
A video decoder may be configured to receive, in response to receiving a first syntax element indicating that reference picture list information is included in a picture header syntax structure, a second syntax element in the picture header syntax structure indicating whether a collocated picture used for temporal motion vector prediction is to be derived from a first reference picture list or a second reference picture list; receive a slice of the video data that refers to the picture header syntax structure; and in response to the slice being a P slice, set a value for a third syntax element associated with the slice to a first value for the third syntax element, with the first value for the third syntax element indicating that the collocated picture used for temporal motion vector prediction is to be derived from the first reference picture list.