Pixel Remapping for Dynamic Video Summarization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video surveillance techniques face challenges in efficiently summarizing video content due to errors in object identification and tracking, leading to disjointed playback and high computational requirements, which affect the preservation of activity co-occurrence and continuity.
Innovation Solution
A cyclical, pixel-based remapping technique is used to create a summary video by generating a background image and comparing source video frames pixel-by-pixel to identify active pixels, remapping them over the background to maintain activity co-occurrence and continuity, allowing for dynamic adjustment of parameters like motion sensitivity and time compression ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional video summarization techniques are used, then video duration is compressed, but object identification errors and tracking errors occur leading to disjointed playback
Solution Approach 1:
The patent replaces conventional object-based tracking mechanisms with a pixel-based remapping mechanism. Instead of identifying and tracking objects through complex recognition algorithms, the system directly remaps pixel positions from source video frames to summary video frames based on geometric transformations, eliminating object identification errors entirely
Solution Approach 2:
The patent creates accurate copies of pixel data from source frames to summary frames through remapping. Each pixel's position and visual information are copied and transformed according to calculated mapping relationships, preserving the original video content's accuracy without relying on object recognition
2Loss of time
If conventional video summarization techniques are used, then video duration is compressed, but computational requirements increase
Solution Approach 1:
The patent segments the video processing task into independent pixel-level operations. Instead of performing complex object recognition and tracking on entire frames, the system divides the work into individual pixel remapping operations that can be processed independently and efficiently, reducing overall computational energy requirements
Solution Approach 2:
The patent extracts only the essential information needed for summarization - pixel position and visual data - and discards unnecessary computational steps such as object identification and tracking. This extraction approach reduces computational energy by focusing only on the critical remapping operation
3Loss of time
If conventional video summarization techniques are used, then video duration is compressed, but activity continuity is lost due to tracking errors
Solution Approach 1:
The patent replaces the mechanical system of object tracking with a direct pixel remapping system. By calculating mapping relationships between source and summary frames and directly remapping pixels, the system maintains activity continuity without the errors inherent in object-based tracking methods
Solution Approach 2:
The patent ensures continuity of useful action by maintaining the spatial and temporal relationships of pixels throughout the remapping process. The systematic approach to calculating and applying mapping relationships preserves the continuous nature of video content, ensuring that activities remain coherent in the summary video
Data Source
AI summary
Cyclical pixel-based remapping techniques are disclosed for efficiently creating and displaying a video summary. Cyclical temporal remapping at the pixel level requires less processing power and eliminates the need for object identification and tracking, allowing for dynamic adjustment of summary video parameters while a user is viewing the summary video. An operator can select an archived camera feeds or video file, specify a time compression ratio or time interval over which to review video content, specify a motion sensitivity threshold, and then view a significantly condensed summary video that depicts all activity from the source video within a much shorter timeline. Furthermore, activity co-occurrence is preserved such that any activities that co-occur in the source video will also co-occur in the summary video.