Video Shot Change Detection Using CDF Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing techniques for detecting shot changes in videos are ineffective in accurately identifying spatial differences between frames with similar luminance and color values, and often generate false positives due to sudden lighting changes, leading to low precision in shot change detection.

Innovation Solution

A method that calculates difference values for pixel blocks in video frames, generates cumulative distribution functions (CDFs) based on these values, and compares them to determine shot changes, allowing for precise detection in both forward and reverse directions, while avoiding false positives from lighting changes within a single shot.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If histogram-based techniques are used to detect shot changes, then the method is simple to implement, but it fails to detect spatial differences between frames with similar luminance and color values

Engineering Contradiction:
Improveease of implementationVSAvoiddetection accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent divides the video frame into multiple blocks and calculates difference values for each block individually. This segmentation allows the system to detect spatial differences between frames by examining local variations, overcoming the limitation of histogram-based methods that treat the entire frame as a single unit and miss spatial changes when overall luminance and color values are similar.

Inventive Principle:
Principle #1Segmentation

2Reliability

If sum of difference values between consecutive video frames is used to detect shot changes, then shot changes are detected with high recall, but false positives increase due to significant lighting changes

Engineering Contradiction:
ImproverecallVSAvoidprecision
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies different analysis approaches to different parts of the video stream by examining blocks individually and comparing their difference value distributions. This local quality approach allows the system to distinguish between genuine shot changes (where most blocks show significant differences) and lighting changes within a single shot (where only certain blocks are affected), thereby reducing false positives while maintaining high recall.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent transforms the raw difference values into a cumulative distribution function (CDF) to analyze the statistical distribution of differences across blocks. This parameter transformation enables the system to detect shot changes by identifying significant shifts in the distribution pattern, rather than relying on absolute difference thresholds that are sensitive to lighting changes, thus improving precision while maintaining recall.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If cumulative distribution function comparison is used to detect shot changes, then precision is improved by identifying spatial differences, but computational complexity increases

Engineering Contradiction:
Improvedetection precisionVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent pre-calculates and stores the cumulative distribution functions for blocks of frames during a first pass through the video. This preliminary action allows the system to have the computationally intensive CDF calculations completed in advance, so that during the actual shot change detection, only comparisons are needed rather than full CDF calculations, thereby reducing real-time computational complexity while maintaining high detection precision.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10686969B2Detecting shot changes in a video
Publication Date: 2020.06.16 NETFLIX INC
  • US10686969B2 patent drawing
  • US10686969B2 patent drawing
  • US10686969B2 patent drawing

AI summary

One embodiment of the present invention sets forth a technique for detecting a shot change in a video. The technique includes calculating a first plurality of difference values for pixel blocks included in a first video frame and a second video frame and calculating a first cumulative distribution function (CDF) based on the first plurality of difference values. The technique further includes calculating a second plurality of difference values for pixel blocks included in a first plurality of video frames that is contiguous with the first video frame and calculating a second CDF based on the second plurality of difference values. The technique further includes comparing the first CDF to the second CDF to detect a shot change.