Video Encoder DPB Output Time Offset Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in optimizing processing efficiency and image quality, particularly in managing temporal sublayers and buffering delays, which can lead to reverse phenomena where pictures are output before they are decoded, resulting in increased processing complexity and code amount.
Innovation Solution
The proposed solution involves an encoder and decoder that calculate and adjust decoded picture buffer (DPB) output times by subtracting DPB output deltas from DPB output delays, using offset values stored in the bitstream, ensuring that DPB output times do not precede coded picture buffer (CPB) removal times, thereby eliminating reverse phenomena and reducing processing and code amounts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If DPB output times are calculated without adjusting for temporal sublayer offsets, then processing is simpler, but reverse phenomena occur where pictures are output before decoded resulting in increased processing complexity
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing offset values for each temporal sublayer in the bitstream during encoding. This allows the decoder to directly retrieve and apply these offsets when calculating DPB output times, eliminating the need for complex runtime calculations and preventing reverse phenomena before they occur.
Solution Approach 2:
The patent changes the parameter representation by introducing separate offset values for each temporal sublayer (id=0, 1, 2, etc.) rather than using a single unified timing parameter. This allows independent adjustment of DPB output times for different temporal sublayers, enabling precise control to prevent pictures from being output before decoded while maintaining processing efficiency.
2Productivity
If separate DPB output delta values are provided for each temporal sublayer, then coding efficiency improves, but code amount increases
Solution Approach 1:
The patent changes the parameter structure by representing DPB output adjustments as offset values relative to a base timing for the highest temporal sublayer. This parameter transformation allows efficient encoding of timing information for multiple temporal sublayers using differential values, improving coding efficiency while controlling code amount through selective precision.
Solution Approach 2:
The patent applies partial action by providing offset values only for temporal sublayers that require adjustment, rather than uniformly encoding timing information for all sublayers. This selective approach maintains coding efficiency for sublayers needing precise control while reducing code amount for sublayers with standard timing relationships.
3Quantity of substance
If DPB output times are not adjusted using offset values, then code amount is reduced, but reverse phenomena occur where pictures are output before decoded
Solution Approach 1:
The patent introduces offset values as intermediary parameters that mediate between the base timing information and the actual DPB output times for each temporal sublayer. These intermediaries enable precise timing control to prevent reverse phenomena while maintaining compact representation, thus preserving both code amount efficiency and decoding reliability.
Solution Approach 2:
The patent applies preliminary action by pre-computing and embedding offset values in the bitstream during encoding, so that the decoder can reliably calculate correct DPB output times without performing complex runtime adjustments. This preliminary preparation ensures decoding reliability while keeping the additional code amount minimal.
Data Source
AI summary
An encoder includes circuitry and memory coupled to the circuitry. Given id indicating a lower temporal sublayer other than a highest temporal sublayer in temporal sublayers, the circuitry calculates a DPB output time [id] for a picture in the lower temporal sublayer. In doing so, the circuitry of the encoder subtracts a DPB output delta [id] provided for each of the temporal sublayers from a DPB output delay shared between the temporal sublayers, and given maxid indicating the highest temporal sublayer, calculates the DPB output delta [id] by subtracting a CPB removal delay [maxid] and an offset [id] from a CPB removal delay [id]. Next, the circuitry stores the offset [id] in a bitstream.