Patents by Inventor Utkarsh MOHAN

Utkarsh MOHAN has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • System and method for ultra low latency live streaming based on user datagram protocol
    Patent number: 12177282
    Abstract: An example method includes for enhancing video streaming includes inputting a client-side environment parameter to a client-side neural network (NN) implemented in a client device and then training the client-side NN based on the inputted client-side environment parameter. Thereafter, a client-side encoding bit rate is determined based on a result of the training of the client-side NN and then transmitting the determined client-side encoding bit rate as an input to a server-side neural network (NN) implemented in a server device. The method further includes training the server-side NN based on the inputted client-side encoding bit rate and a server-side environment parameter. Then, a server-side encoding bit rate is determined based on a result of the training of the server-side NN and thereafter, an encoding bit rate of an encoder for the video stream is adjusted based on the determination of the server-side encoding bit rate, and the server-side environment parameter.
    Type: Grant
    Filed: September 28, 2022
    Date of Patent: December 24, 2024
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Luckraj Shrawan Kumar, Gopinath Chennakeswaran, Renuka Varry, Divyanshu Chuchra, Umang Bhatia, Utkarsh Mohan, Aalok Kumar Gupta
  • SYSTEM AND METHOD FOR ULTRA LOW LATENCY LIVE STREAMING BASED ON USER DATAGRAM PROTOCOL
    Publication number: 20230094815
    Abstract: An example method includes for enhancing video streaming includes inputting a client-side environment parameter to a client-side neural network (NN) implemented in a client device and then training the client-side NN based on the inputted client-side environment parameter. Thereafter, a client-side encoding bit rate is determined based on a result of the training of the client-side NN and then transmitting the determined client-side encoding bit rate as an input to a server-side neural network (NN) implemented in a server device. The method further includes training the server-side NN based on the inputted client-side encoding bit rate and a server-side environment parameter. Then, a server-side encoding bit rate is determined based on a result of the training of the server-side NN and thereafter, an encoding bit rate of an encoder for the video stream is adjusted based on the determination of the server-side encoding bit rate, and the server-side environment parameter.
    Type: Application
    Filed: September 28, 2022
    Publication date: March 30, 2023
    Inventors: Luckraj Shrawan KUMAR, Gopinath CHENNAKESWARAN, Renuka VARRY, Divyanshu CHUCHRA, Umang BHATIA, Utkarsh MOHAN, Aalok Kumar GUPTA