Video Frame Type Classification Using Accumulation Matrix Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for determining video frame type, such as stereoscopic and monoscopic frames, require supplementary data or specific decoding capabilities, which can be problematic for systems lacking these functionalities or when video signals lack explicit type indication.

Innovation Solution

A computer-implemented method and system that analyzes video frames by extracting active regions, filtering and comparing characteristic features between frame halves to classify frames as 2D or 3D, using a frame compactor and detector to identify non-active regions and calculate accumulation matrices for determining frame type.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If supplementary data or signals are used to indicate video frame type, then the accuracy of frame type determination is improved, but the device complexity and requirement for decoding capabilities increase

Engineering Contradiction:
Improveframe type determination accuracyVSAvoiddecoding capability requirement
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The video frame itself contains all necessary information for type determination through its intrinsic structure. The method analyzes the frame's active region distribution, pixel correlation, and structural patterns without requiring external supplementary data or complex decoding capabilities. The frame determines its own type through self-contained structural analysis.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The method extracts and analyzes only the essential structural features directly present in the video frame, such as active region locations, pixel correlations between frame halves, and structural patterns. This extraction approach eliminates the need for supplementary data while focusing on the most discriminative frame characteristics.

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If supplementary data or signals are used to indicate video frame type, then the reliability of signal type recognition is improved, but the ease of operation and compatibility decrease

Engineering Contradiction:
Improvesignal type recognition accuracyVSAvoidsystem compatibility
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The method provides a universal solution that works across different video formats, display devices, and signal types. By analyzing intrinsic frame structures rather than relying on format-specific supplementary data, the system achieves broad compatibility while maintaining reliable frame type determination for various stereoscopic and monoscopic formats.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The video frame itself provides all necessary information for reliable type recognition through its inherent structural properties. The method extracts type indicators directly from the frame content, making the system reliable without requiring external supplementary signals or complex decoding capabilities.

Inventive Principle:
Principle #25Self-service

3Ease of operation

If basic image content without type indication is analyzed, then the ease of operation and compatibility are improved, but the difficulty of detecting and measuring frame type increases

Engineering Contradiction:
Improvesystem compatibilityVSAvoidframe type detection complexity
Core Design Contradiction:
Ease of operationVSDifficulty of detecting and measuring

Solution Approach 1:

The method focuses analysis on specific local structural qualities that differentiate frame types, such as the distribution and positioning of active regions, local pixel correlations within frame halves, and structural patterns in specific areas. This localized quality analysis simplifies detection by concentrating on discriminative features rather than analyzing the entire frame uniformly.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The video frame is segmented into distinct regions (first half and second half, active and non-active regions) to facilitate systematic comparison. This segmentation enables the detection method to analyze specific portions independently and combine results, reducing overall detection complexity while improving accuracy.

Inventive Principle:
Principle #1Segmentation

4Measurement precision

If manual specification of video frame type is used, then the accuracy of frame type indication is improved, but the ease of operation decreases due to user burden

Engineering Contradiction:
Improveframe type indication accuracyVSAvoiduser operation simplicity
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system automatically determines frame type by analyzing intrinsic structural properties of the video frame itself, eliminating the need for manual user specification. The frame's own structure provides the information needed for accurate type determination, making the process both accurate and automated.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The method uses changes in structural parameters (active region distribution, pixel correlation values, structural patterns) to automatically infer frame type. By monitoring these parameter variations, the system achieves accurate frame type determination without requiring manual user input.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2477407B1A method and a system for determining a video frame type
Publication Date: 2018.03.21 ADVANCED DIGITAL BROADCAST
  • EP2477407B1 patent drawingFigure 1A~1D
  • EP2477407B1 patent drawingFigure 2A~2H
  • EP2477407B1 patent drawingFigure 3A~3H

AI summary

A computer-implemented method for determining a video frame type, comprising the steps of receiving a video frame, filtering the video frame to enhance characteristic features (AL-GL, AR-GR) of the contents of the video frame, creating a list of coordinates of at least part of the characteristic features for a first half of the frame (AL-GL) and the other half (AR-GR) of the frame, comparing the coordinates (i,j) of the characteristic features of the list for the first half of the frame with the coordinates (i,j) of the characteristic features of the other half of the frame to obtain an accumulation matrix (UΔi,Δj, VΔi,Δj) having a peak value (xLR, xTB) specifying the most frequent coordinate difference and a mean value (mLR, mTB) specifying the mean coordinate difference and classifying the frame as a stereoscopic frame if the peak value (xLR, xTB) exceeds the mean value (mLR, mTB) by a predefined threshold and as a monoscopic frame otherwise.