Video Analysis Method Using Packet Size Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting motion in real-time video streaming services require high computing power and are codec-dependent, limiting their ability to process videos in real-time and apply motion detection to both transmitting and receiving ends effectively.
Innovation Solution
A method and apparatus that analyze the pattern of network packet sizes to determine motion, eliminating the need for video decoding and reducing computing power requirements by calculating frame sizes and recognizing frames without decoding, allowing for motion detection based on frame size patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video decoding and image comparison methods are used for motion detection, then motion detection accuracy is improved, but computing power requirements increase and real-time processing capability deteriorates
Solution Approach 1:
The patent extracts only the essential information needed for motion detection from the video data. Instead of decoding and processing complete video frames, the system extracts motion detection results by analyzing packet size patterns and frame size variations, which contain sufficient information to determine motion occurrence without requiring full video decoding and image comparison operations.
Solution Approach 2:
The patent inverts the conventional motion detection approach by not directly analyzing video content for motion. Instead, it analyzes the metadata characteristics (packet sizes, frame sizes) of the video stream, which indirectly reflect motion information. This inversion allows motion detection without the computationally expensive steps of video decoding and pixel-level image comparison.
2Adaptability or versatility
If video decoding is performed for motion detection, then motion detection capability is improved, but codec dependency increases and device compatibility deteriorates
Solution Approach 1:
The patent introduces packet size analysis and frame size analysis as intermediary methods between the video stream and motion detection. These intermediaries operate at the network packet level rather than requiring video decoding, serving as a codec-agnostic bridge that enables motion detection across different video codecs without requiring device-specific decoding capabilities.
3Speed
If real-time video processing is performed at receiving terminals, then motion detection speed is improved, but network data transmission volume increases
Solution Approach 1:
The patent applies partial action by performing only the necessary analysis for motion detection without processing the complete video data. Instead of receiving and processing full video frames, the system analyzes packet size patterns and frame size variations, which are partial data representations that contain sufficient information for motion detection while significantly reducing network data transmission requirements.
Data Source
AI summary
Provided is a video analysis (VA) method comprises calculating, by a VA apparatus, a size of each of a plurality of frames by summing sizes of a plurality of packets constituting each of the plurality of frames, analyzing, by the VA apparatus, the size of each of the plurality of frames constituting a video to determine a pattern of the sizes of the plurality of frames and determining, by the VA apparatus, whether there is a motion in the video based on the size pattern.