Method and Apparatus for Video Compression Using Skip Modes

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention relates to a video compression method and apparatus that uses advanced 'skip modes' to reduce the amount of data that needs to be encoded and transmitted. The core idea is to analyze motion in different regions of each video frame and then decide, based on the estimated motion (using measures like motion variance or entropy), which parts of the data are redundant enough to be safely skipped during encoding. These skipped parts are not transmitted, reducing the total bitrate, while preserving important information in areas with more motion. The method is particularly optimized to work with neural-network-based video codecs.

Use CasesContent extracted from patent full text and abstract with AI.

• Streaming video services and platforms to deliver content with reduced bandwidth requirements.
• Video conferencing tools that require real-time video transmission over variable quality networks.
• Video storage systems where large archives of surveillance or security footage need space-saving compression.
• Edge and cloud-based video analytics, enabling efficient transmission of video data from IoT or mobile devices for machine vision tasks without overwhelming bandwidth.
• Mobile devices (e.g., smartphones, tablets, action cameras) for efficient on-device video encoding and sharing.
• Remote monitoring and industrial IoT solutions relying on video feeds, where reducing data usage is critical.

BenefitsContent extracted from patent full text and abstract with AI.

• Significant reduction in the required bitrate for video transmission and storage, saving bandwidth and storage costs.
• Adaptive skipping based on motion ensures that important video regions are retained at higher quality, while static or low-motion areas are compressed more aggressively.
• Faster video decoding and lower computational load due to less data being processed and transmitted.
• Integration with neural network-based codecs allows end-to-end optimized, intelligent video compression suited for both human viewing and machine vision.
• Flexible, extensible design supporting various measures of motion complexity and easy adaptation to different coding scenarios and hardware.
• Supports real-time applications by reducing the data transmission bottleneck, improving user experience in live streaming and conferencing.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

Disclosed is a video compression device (700) having one or more processors (701) configured to execute video compression by: receiving (1201) video data representing a video stream; for each of a plurality of regions in a state of the video stream, estimating (1202) motion between that region in that state and corresponding regions in other states of the video stream; selecting (1203) one or more parts in the state of the video stream as a part to be skipped in dependence on the respective estimated motion for an associated region in the state of the video stream; and encoding (1204) the video data to form a residual representing the state of the video stream in one or more channels by, in response to a part having been selected to be skipped, omitting data representing that part in at least one of the channels from the residual. A corresponding method is also disclosed. This may allow video compression to be performed at reduced bitrate cost.