EFFICIENT WARPING-BASED NEURAL VIDEO CODEC

Organization Name

qualcomm incorporated

Inventor(s)

Ties Jehan Van Rozendaal of Amsterdam (NL)

Hoang Cong Minh Le of Santee CA (US)

Tushar Singhal of San Diego CA (US)

Amir Said of San Diego CA (US)

Krishna Buska of San Diego CA (US)

Guillaume Konrad Sautiere of Amsterdam (NL)

Anjuman Raha of San Diego CA (US)

Auke Joris Wiggers of Amsterdam (NL)

Frank Steven Mayer of San Diego CA (US)

Liang Zhang of San Diego CA (US)

Abhijit Khobare of San Diego CA (US)

Muralidhar Reddy Akula of San Diego CA (US)

EFFICIENT WARPING-BASED NEURAL VIDEO CODEC - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240305785 titled 'EFFICIENT WARPING-BASED NEURAL VIDEO CODEC

The abstract describes a computing device that can decode video data using parallel entropy decoding, motion vector prediction, and block-based warp functions.

The device decodes encoded video data using parallel entropy decoding.
It predicts motion vectors based on the decoded data.
It decodes motion vector residuals from the decoded data.
It combines motion vector residuals with motion vectors.
It uses block-based warp functions to generate predicted current video data.
It sums the predicted data with residual blocks to generate current reconstructed video data.

Potential Applications: This technology can be used in video decoding systems, streaming services, video editing software, and virtual reality applications.

Problems Solved: This technology addresses the challenges of efficiently decoding and reconstructing video data in real-time.

Benefits: The benefits of this technology include improved video quality, reduced latency in video streaming, and enhanced user experience in virtual reality environments.

Commercial Applications: Title: Advanced Video Decoding Technology for Enhanced User Experience This technology can be commercialized in video streaming services, virtual reality platforms, video editing software, and surveillance systems.

Prior Art: Prior research in video compression and decoding techniques, motion estimation algorithms, and block-based warp functions can provide valuable insights into the development of this technology.

Frequently Updated Research: Researchers are continually exploring new algorithms and optimizations to enhance video decoding efficiency and quality. Stay updated on the latest advancements in video compression and decoding technologies.

Questions about Video Decoding Technology: 1. How does parallel entropy decoding improve video decoding efficiency? Parallel entropy decoding allows for faster processing of encoded video data, leading to quicker decoding and reconstruction of video frames. 2. What are the key differences between motion vector prediction and motion vector residuals in video decoding? Motion vector prediction estimates the motion of objects in a video frame, while motion vector residuals represent the difference between predicted and actual motion vectors, helping to refine the motion estimation process.

Original Abstract Submitted

an example computing device may include memory and one or more processors. the one or more processors may be configured to parallel entropy decode encoded video data from a received bitstream to generate entropy decoded data. the one or more processors may be configured to predict a motion vector based on the entropy decoded data. the one or more processors may be configured to decode a motion vector residual from the entropy decoded data. the one or more processors may be configured to add the motion vector residual and motion vector. the one or more processors may be configured to warp previous reconstructed video data with an overlapped block-based warp function using the motion vector to generate predicted current video data. the one or more processors may be configured to sum the predicted current video data with a residual block to generate current reconstructed video data.