View source for Apple inc. (20240163436). JUST NOTICEABLE DIFFERENCES-BASED VIDEO ENCODING simplified abstract

=JUST NOTICEABLE DIFFERENCES-BASED VIDEO ENCODING=


==Organization Name==

[[:Category:apple inc.|apple inc.]]


[[Category:apple inc.]]

==Inventor(s)==

[[:Category:Wei Li of Saratoga CA (US)|Wei Li of Saratoga CA (US)]][[Category:Wei Li of Saratoga CA (US)]]

[[:Category:Hye-Yeon Cheong of Los Gatos CA (US)|Hye-Yeon Cheong of Los Gatos CA (US)]][[Category:Hye-Yeon Cheong of Los Gatos CA (US)]]

[[:Category:Jiancong Luo of San Diego CA (US)|Jiancong Luo of San Diego CA (US)]][[Category:Jiancong Luo of San Diego CA (US)]]

[[:Category:Linfeng Guo of Cupertino CA (US)|Linfeng Guo of Cupertino CA (US)]][[Category:Linfeng Guo of Cupertino CA (US)]]

==JUST NOTICEABLE DIFFERENCES-BASED VIDEO ENCODING - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20240163436 titled 'JUST NOTICEABLE DIFFERENCES-BASED VIDEO ENCODING

==Simplified Explanation==

The patent application describes techniques for achieving quantization in video coding applications that result in high coding efficiency and maintain high image quality. These techniques utilize quantization parameters based on just noticeable difference (JND) models to estimate coding artifacts in video coding. The process involves predictively coding an input pixel block with reference to a prediction reference, transforming prediction residuals into transform domain coefficients, and quantizing transform coefficients using JND-quality quantization values from a table indexed by statistical analysis of the input pixel block.

* Predictive coding of input pixel blocks with a prediction reference
* Transformation of prediction residuals into transform domain coefficients
* Quantization of transform coefficients using JND-quality quantization values
* Utilization of quantization parameters based on JND models for estimating coding artifacts

== Potential Applications ==
The technology can be applied in video coding, streaming services, video surveillance systems, and video conferencing solutions.

== Problems Solved ==
1. Improved coding efficiency in video applications
2. Maintaining high image quality during quantization processes

== Benefits ==
1. Higher coding efficiency
2. Retention of high image quality
3. Enhanced video compression capabilities

== Potential Commercial Applications ==
1. Video streaming platforms
2. Video surveillance companies
3. Telecommunication companies for video conferencing solutions

== Possible Prior Art ==
Previous methods of quantization in video coding applications may not have utilized JND models for estimating coding artifacts, resulting in lower coding efficiency and potential loss of image quality.

=== Unanswered Questions ===

=== How does this technology compare to existing video coding techniques? ===
This article does not provide a direct comparison to existing video coding techniques in terms of coding efficiency and image quality preservation.

=== What are the specific statistical analyses used to index the JND-quality quantization values? ===
The article does not delve into the specific statistical analyses employed to index the JND-quality quantization values in the table.


==Original Abstract Submitted==

techniques are disclosed for achieving quantization in video coding applications that achieves high coding efficiency and retains high image quality. these techniques employ quantization processes using quantization parameters that have been developed according to just noticeable difference (“jnd”) models for estimating coding artifacts from video coding. according to these techniques, an input pixel block of video is predictively coded with reference to a prediction reference, and prediction residuals obtained therefrom are transformed to transform domain coefficients. a transform coefficient is quantized by a quantization parameter read from a table populated by jnd-quality quantization values, which is indexed by a value representing a statistical analysis of the input pixel block.



[[Category:H04N19/124]]
[[Category:H04N19/105]]
[[Category:H04N19/176]]
[[Category:H04N19/61]]