Video Decoder POC Calculation Using Layer Identity References

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In video encoding/decoding processes, particularly in H.264 and HEVC standards, determining the Picture Order Count (POC) of a current picture is challenging due to the presence of multiple temporal layers, which can result in different POC values for the same picture across decoders, leading to instability in the decoding process.

Innovation Solution

The POC of a current picture is calculated as a sum of its least significant bits and most significant bits, with the most significant bits derived from a previous reference picture's most significant bits and least significant bits, ensuring that the previous reference picture has a layer identity equal to or lower than the current picture's layer identity, thereby stabilizing the POC calculation regardless of the number of decoded temporal layers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If POC is calculated using previous reference picture values in multi-temporal-layer decoding, then decoding continuity is maintained, but POC values become inconsistent across different decoders

Engineering Contradiction:
Improvedecoding stabilityVSAvoidPOC value consistency
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The POC value is segmented into two parts: least significant bits (LSB) signaled in the bitstream and most significant bits (MSB) derived by the decoder. This segmentation allows the LSB to carry picture-specific information while the MSB is derived from a fixed reference (temporal_id=0 picture), ensuring consistency across decoders. The formula PicOrderCnt=PicOrderCntMsb+pic_order_cnt_Isb separates the stable base value from the variable offset.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention extracts the temporal layer identifier (temporal_id) as a separate parameter to determine which reference picture should be used for POC derivation. By explicitly checking temporal_id equality, the system extracts only the appropriate reference picture values that are compatible with the current picture's temporal layer, preventing inconsistent derivations from pictures in different temporal layers.

Inventive Principle:
Principle #2Taking out (Extraction)

2Adaptability or versatility

If multiple temporal layers are decoded, then video scalability is improved, but POC calculation becomes ambiguous

Engineering Contradiction:
Improvetemporal scalabilityVSAvoidPOC calculation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The invention applies local quality by treating different temporal layers differently in the POC derivation process. Pictures with temporal_id=0 serve as the fixed reference point with explicit POC signaling, while pictures with higher temporal_id values derive their POC from these base pictures. This local differentiation ensures that each temporal layer has a well-defined reference hierarchy, simplifying the overall calculation despite multi-layer complexity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system performs preliminary action by establishing the temporal_id=0 picture as the foundational reference before deriving POC values for higher temporal layers. The POC values of temporal_id=0 pictures are explicitly signaled and stored as prevPicOrderCntMsb and prevPicOrderCntLsb, creating a stable base that subsequent layers can reliably reference without recalculation or ambiguity.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If POC is derived from any previous reference picture, then decoding efficiency is improved, but display order becomes incorrect

Engineering Contradiction:
Improvedecoding efficiencyVSAvoiddisplay order accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The invention implements feedback by continuously monitoring the temporal_id of reference pictures during the POC derivation process. The condition temporal_id equal to or lower than the current picture's temporal_id acts as a feedback mechanism that ensures only compatible reference pictures are used. This feedback loop prevents incorrect POC assignments while maintaining efficient decoding by allowing valid reference picture utilization.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20240380912A1Decoders and Methods Thereof for Managing Pictures in Video Decoding Process
Publication Date: 2024.11.14 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • US20240380912A1 patent drawing
  • US20240380912A1 patent drawing
  • US20240380912A1 patent drawing

AI summary

The solution presented herein defines from which previously decoded picture to retrieve the values for calculating POC of the current picture regardless of how many temporal layers have been decoded. That is achieved by determining the POC of the current picture, to be used by the decoder, as a sum of a syntax element pic_order_cnt_Isb and most significant bits of the POC, PicOrderCntMsb, of the current picture, wherein the PicOrderCntMsb of the current picture is derived using at least a prevPicOrderCntMsb and a prevPicOrderCntLsb and the prevPicOrderCntMsb is set equal to the most significant bits of the POC of a previous reference picture in decoding order that has a layer identity equal to zero and prevPicOrderCntLsb is set equal to the value of the least significant bits of the POC of a previous reference picture in decoding order that has a layer identity equal to zero.