Video Summary Generation via Real-Time Feature Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing digital video processing systems struggle to generate video summaries on digital capture devices with minimal delay, especially when users want to include specific features like people, pets, events, or objects, as existing automatic summarization algorithms require decompressing the video data, making it impractical for quick review and sharing.

Innovation Solution

A digital video camera system that captures video sequences and generates a summary by analyzing image frames in real-time using a feature recognition algorithm, storing the summary in metadata without decompressing the video data, allowing for immediate viewing and sharing on the device.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If automatic video summarization algorithms are used, then video summary generation is automated, but the process requires decompressing video data which causes significant delay

Engineering Contradiction:
Improvevideo summary generationVSAvoidtime delay
Core Design Contradiction:
Extent of automationVSLoss of time

Solution Approach 1:

The system performs preliminary actions by extracting and analyzing key frames during the video capture process itself, rather than waiting for the complete video to be recorded and then processing it. This allows the summarization to be prepared in advance as the video is being captured, significantly reducing the time delay when the user needs to review the summary.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The video processing is segmented into discrete key frames that are analyzed independently. Instead of processing the entire compressed video stream, the system extracts specific representative frames (key frames) that capture important moments, processes only these segments, and reconstructs the summary from them. This segmentation dramatically reduces processing time and data requirements.

Inventive Principle:
Principle #1Segmentation

2Quantity of substance

If video data is compressed to reduce size, then sharing becomes more practical, but analysis requires decompression which increases processing time

Engineering Contradiction:
Improvevideo data sizeVSAvoidprocessing time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The system extracts only the essential information (key frames) from the compressed video data without needing to decompress the entire video stream. By taking out and analyzing only these representative frames, the system can perform summarization analysis on compressed data directly, avoiding the time-consuming decompression step while still obtaining meaningful summary results.

Inventive Principle:
Principle #2Taking out (Extraction)

3Ease of operation

If manual video editing is used to create summaries, then user control over content is high, but the process is lengthy and laborious

Engineering Contradiction:
Improveuser controlVSAvoidediting time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system provides self-service by automatically analyzing the video content and generating summaries without requiring user intervention. The automated algorithm independently identifies key moments, selects representative frames, and constructs the video summary, eliminating the need for manual editing while giving users control over parameters like feature detection sensitivity and summary length.

Inventive Principle:
Principle #25Self-service

4Adaptability or versatility

If feature-specific video summaries are created, then relevance to user needs is improved, but the complexity of analysis increases

Engineering Contradiction:
Improvefeature-specific summarizationVSAvoidanalysis complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system applies local quality by detecting and analyzing specific features of interest (such as particular objects, people, or actions) within the video frames. Instead of uniformly processing the entire video, the feature detection algorithm identifies and focuses analysis only on frames containing these specific features, making the summary tailored to user needs while managing complexity through targeted rather than comprehensive analysis.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2710594B1Video summary including a feature of interest
Publication Date: 2016.07.20 INTELLECTUAL VENTURES FUND 83 LLC
  • EP2710594B1 patent drawingFigure 1
  • EP2710594B1 patent drawingFigure 2
  • EP2710594B1 patent drawingFigure 3

AI summary

A digital video camera system that provides a video summary using a method that includes: specifying reference data, wherein the reference data indicates a feature of interest; capturing a video sequence of the scene using the image sensor, the video sequence including a time sequence of image frames; processing the captured video sequence using a video processing path to form a digital video file; during the capturing of the video sequence, analyzing the captured image frames using a feature recognition algorithm to identify a subset of the image frames that contain the feature of interest; forming the video summary including fewer than all of the image frames in the captured video sequence, wherein the video summary includes at least part of the identified subset of image frames containing the feature of interest; storing the digital video file in the storage memory; and storing a representation of the video summary in the storage memory.