Video Summary Generation by Extracting and Overlaying Event Pixels

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Monitoring and reviewing video recordings is a time-consuming task, especially when searching for specific events without knowing the exact time, as increasing playback speed or using multiple windows can lead to missing important details or becoming overwhelmed with multiple streams.

Innovation Solution

Generating a summary video sequence by identifying event video sequences with objects of interest, extracting and overlaying their pixels while maintaining spatial and temporal relationships, and disregarding time periods without these objects, to create a condensed version that preserves concurrent actions and interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If playback speed is increased to reduce review time, then time efficiency is improved, but important details may be missed

Engineering Contradiction:
Improvereview timeVSAvoidimportant details
Core Design Contradiction:
Loss of timeVSLoss of information

Solution Approach 1:

The patent extracts only the relevant portions of video containing objects of interest from the complete video sequence. By identifying and isolating event video sequences where objects of interest are present, the system creates a condensed summary that eliminates irrelevant time periods while preserving all important information, allowing reviewers to watch at normal speed without missing details.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the complete video sequence into multiple event video sequences based on the presence and absence of objects of interest. Each segment contains only relevant content with objects of interest, and these segments are concatenated to form a compressed summary sequence, enabling efficient review without information loss.

Inventive Principle:
Principle #1Segmentation

2Productivity

If multiple windows are used to view different video portions simultaneously, then review speed is improved, but complexity of tracking events increases

Engineering Contradiction:
Improvereview speedVSAvoidtracking complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent merges multiple event video sequences containing objects of interest into a single compressed summary sequence. By overlaying and concatenating relevant segments from different time periods into one continuous video stream, the system allows reviewers to view all important events in a single window, eliminating the need to track multiple simultaneous streams while maintaining review efficiency.

Inventive Principle:
Principle #5Merging (Combining)

3Loss of time

If playback speed is increased by skipping frames, then review time is reduced, but reliability of event detection decreases

Engineering Contradiction:
Improvereview timeVSAvoidevent detection reliability
Core Design Contradiction:
Loss of timeVSReliability

Solution Approach 1:

The patent extracts complete event video sequences containing objects of interest at their original playback speed from the compressed video sequence. By preserving the full frame sequences of relevant events rather than using accelerated or skipped frames, the system ensures high reliability in event detection while still achieving time savings by excluding irrelevant time periods from the summary.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS10192119B2Generating a summary video sequence from a source video sequence
Publication Date: 2019.01.29 AXIS
  • US10192119B2 patent drawing
  • US10192119B2 patent drawing
  • US10192119B2 patent drawing

AI summary

A method for generating a summary video sequence from a source video sequence is disclosed. The method comprises: identifying, in the source video sequence, event video sequences, wherein each event video sequence comprises consecutive video frames in which one or more objects of interest are present; extracting, from video frames of one or more event video sequences of the event video sequences, pixels depicting the respective one or more objects of interest; while keeping spatial and temporal relations of the extracted pixels as in the source video sequence, overlaying the extracted pixels of the video frames of the one or more event video sequences onto video frames of a main event video sequence of the event video sequences, thereby generating the summary video sequence. A video processing device configured to generate the summary video sequence is also disclosed.