20240015312. NEURAL NETWORK BASED FILTERING PROCESS FOR MULTIPLE COLOR COMPONENTS IN VIDEO CODING simplified abstract (QUALCOMM Incorporated)

From WikiPatents
Jump to navigation Jump to search

NEURAL NETWORK BASED FILTERING PROCESS FOR MULTIPLE COLOR COMPONENTS IN VIDEO CODING

Organization Name

QUALCOMM Incorporated

Inventor(s)

Hongtao Wang of San Diego CA (US)

Samuel James Eadie of Munchen (DE)

Muhammed Zeyd Coban of Carlsbad CA (US)

Marta Karczewicz of San Diego CA (US)

NEURAL NETWORK BASED FILTERING PROCESS FOR MULTIPLE COLOR COMPONENTS IN VIDEO CODING - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240015312 titled 'NEURAL NETWORK BASED FILTERING PROCESS FOR MULTIPLE COLOR COMPONENTS IN VIDEO CODING

Simplified Explanation

The abstract of this patent application describes a method for processing video data using a neural network model. The method involves receiving a syntax element that defines a filtering mode for the neural network model, which is applied to both a first color component and a second color component. The model is then used to process a first block of the first color component, resulting in a first filtered block that is stored for a coding unit.

  • The method involves processing video data using a neural network model.
  • A syntax element is received to define the filtering mode for the model.
  • The model is applied to both a first color component and a second color component.
  • The model is used to process a first block of the first color component.
  • The result is a first filtered block that is stored for a coding unit.

Potential applications of this technology:

  • Video compression: The method can be used in video compression algorithms to improve the efficiency of encoding and decoding processes.
  • Image enhancement: The neural network model can be applied to enhance the quality of images or videos by filtering out noise or improving sharpness.
  • Object recognition: The method can be utilized in object recognition systems to improve the accuracy and speed of identifying objects in videos.

Problems solved by this technology:

  • Improved video quality: By applying the neural network model in the defined filtering mode, the method can enhance the quality of video data by reducing noise and artifacts.
  • Efficient video processing: The use of the neural network model allows for faster and more efficient processing of video data compared to traditional methods.
  • Accurate object recognition: The method can help improve the accuracy of object recognition systems by applying the neural network model to extract relevant features from video data.

Benefits of this technology:

  • Higher compression efficiency: By utilizing the neural network model, the method can achieve higher compression ratios without significant loss in video quality.
  • Real-time processing: The efficiency of the method allows for real-time video processing, making it suitable for applications that require immediate results.
  • Improved visual experience: The application of the neural network model can enhance the visual experience of viewers by reducing noise and improving the overall quality of videos.


Original Abstract Submitted

a method of processing video data includes receiving a syntax element that defines a filtering mode for a neural network (nn) model for both a first color component and a second color component, applying an instance of the nn model, in the defined filtering mode, to a first block of the first color component to generate a first filtered block, and storing the first filtered block for a coding unit (cu).

Retrieved from "http://wikipatents.org/index.php?title=20240015312._NEURAL_NETWORK_BASED_FILTERING_PROCESS_FOR_MULTIPLE_COLOR_COMPONENTS_IN_VIDEO_CODING_simplified_abstract_(QUALCOMM_Incorporated)&oldid=27906"