Bypass Sub-Stream Entropy Encoding for Low-Latency Bitstreams

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Entropy encoding and decoding processes, such as arithmetic encoding, can be bottlenecks for processing image, video, and point cloud data due to inefficiencies in compressing symbols with even probabilities and the high computational resources required for non-compressible symbols, leading to increased latency and processing time.

Innovation Solution

A system that separates bitstreams into compressible and non-compressible symbol sub-streams, where only the compressible symbols are entropy encoded, and the non-compressible symbols are bypassed, allowing for the formation of chunks or packets that can be transmitted or stored more efficiently, reducing latency by allowing partial bitstreams to be processed and transmitted without waiting for full bitstream processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If all symbols are entropy encoded, then compression efficiency is improved, but encoding latency and processing time increase

Engineering Contradiction:

Improvecompression efficiencyVSAvoidencoding latency

Core Design Contradiction:

Loss of substanceVSLoss of time

Solution Approach 1:

The patent segments the bitstream into two separate sub-streams: a compressible symbol sub-stream that undergoes entropy encoding and a non-compressible symbol sub-stream that bypasses the encoder. This segmentation allows the system to apply compression only where beneficial while avoiding the latency penalty of entropy encoding symbols with even probabilities, thus resolving the contradiction between compression efficiency and encoding latency.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by selectively entropy encoding only the compressible symbol sub-stream rather than all symbols. The non-compressible symbol sub-stream is excluded from entropy encoding, performing the action partially rather than completely. This reduces overall encoding latency while maintaining compression efficiency for the portion of data that benefits from it.

Inventive Principle:

Principle #16Partial or excessive action

2Loss of substance

If entropy encoding is applied to all symbols, then data compression is improved, but computational resource requirements increase

Engineering Contradiction:

Improvedata compressionVSAvoidcomputational resource requirements

Core Design Contradiction:

Loss of substanceVSUse of energy by moving object

Solution Approach 1:

The patent divides the input bitstream into compressible and non-compressible symbol sub-streams, applying entropy encoding only to the compressible portion. This segmentation reduces computational resource requirements by excluding symbols with even probabilities from the resource-intensive entropy encoding process, while still achieving data compression for the beneficial portion of the data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the processing parameter by creating a bypass path for non-compressible symbols that alters their processing state. Instead of undergoing entropy encoding, these symbols are directly transmitted through the bypass sub-stream, changing the computational parameters from high-resource entropy encoding to low-resource direct transmission, thus reducing overall computational resource requirements.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If full bitstreams are processed before transmission, then encoding completeness is ensured, but transmission latency increases

Engineering Contradiction:

Improveencoding completenessVSAvoidtransmission latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by separating symbols into compressible and non-compressible sub-streams before the entropy encoding stage. This preliminary segmentation allows the non-compressible sub-stream to be prepared for transmission independently and concurrently with the entropy encoding of the compressible sub-stream, enabling earlier transmission readiness without compromising encoding completeness.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent maintains continuity of useful action by allowing the non-compressible symbol sub-stream to flow continuously through the bypass path without being blocked by the entropy encoding process. This continuous flow enables transmission to begin as soon as the separation is complete, rather than waiting for full bitstream processing, thus reducing transmission latency while ensuring encoding completeness through proper chunk formation.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS11705924B2Low-latency encoding using a bypass sub-stream and an entropy encoded sub-stream

Publication Date: 2023.07.18 APPLE INC

AI summary

A system comprises an encoder configured to entropy encode a bitstream comprising both compressible and non-compressible symbols. The encoder parses the bitstream into a compressible symbol sub-stream and a non-compressible sub-stream. The non-compressible symbol sub-stream bypass an entropy encoding component of the encoder while the compressible symbol sub-stream is entropy encoded. When a quantity of bytes of entropy encoded symbols and bypass symbols is accumulated a chunk of fixed or known size is formed using the accumulated entropy encoded symbol bytes and the bypass bytes without waiting on the full bitstream to be processed by the encoder. In a complementary manner, a decoder reconstructs the bitstream from the packets or chunks.