Film Grain Synthesis Using Sample-Based Encoding Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional video encoding methods struggle to preserve film grain in motion pictures due to its random nature, often requiring high bitrates to maintain quality, which contradicts the goal of compression efficiency, and the synthesis process increases decoding complexity.
Innovation Solution
A method that determines film grain synthesis parameters based on characteristics of the current sample, such as shape, location, and encoding bits, to apply film grain values weighted by specific factors, reducing the complexity of the synthesis process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional video encoding methods are used to preserve film grain, then film grain quality is maintained, but bitrate increases significantly
Solution Approach 1:
The patent applies preliminary action by extracting film grain characteristics from the original video signal before encoding, storing them as separate metadata, and then synthesizing film grain during decoding. This separates the film grain preservation process from the main video encoding process, allowing grain to be maintained without increasing the bitrate allocated for video data.
Solution Approach 2:
The patent introduces an intermediary approach by using film grain metadata as a mediator between the encoded video signal and the final decoded output. Instead of directly encoding film grain with the video data, the metadata serves as an intermediate representation that guides the synthesis process during decoding, efficiently preserving grain characteristics.
2Adaptability or versatility
If film grain parameters are extracted from reconstructed pictures, then film grain adaptation to picture content is achieved, but decoding complexity increases
Solution Approach 1:
The patent applies preliminary action by extracting film grain characteristics during the encoding stage and storing them in metadata, so that during decoding the film grain synthesis can directly use these pre-extracted characteristics without needing to perform complex extraction operations on reconstructed pictures.
Solution Approach 2:
The patent uses copying by creating a simplified representation of film grain characteristics in the metadata during encoding, which is then copied and applied during decoding. This avoids the need for complex real-time analysis during decoding while maintaining adaptability to picture content.
3Reliability
If high bitrates are used to compress video, then film grain is preserved during encoding, but compression efficiency decreases
Solution Approach 1:
The patent applies segmentation by separating film grain information from the main video signal and encoding them independently. The video signal is compressed using standard techniques while film grain characteristics are extracted and stored as separate metadata, allowing both to be optimized independently for their respective requirements.
Solution Approach 2:
The patent introduces film grain metadata as an intermediary that carries grain characteristics separately from the compressed video data. This intermediary structure allows the video to be highly compressed while grain information is preserved in a compact metadata format, maintaining compression efficiency.
Data Source
AI summary
A method comprising obtaining (601) at least one characteristic of a current sample of a picture before reconstructing the current sample; determining (602) parameters of a film grain synthesis process to be applied to the current sample from the at least one characteristic; storing (603) the determined parameters with information representative of a location of the current sample; reconstructing (604) the current sample; and, applying (605) the film grain synthesis process on the current sample with the determined parameters.