Курсы английского
<<  METHODS FOR DETECTION AND REMOVAL OF PARASITIC FREQUENCY MODULATION IN AUDIO RECORDINGS Computer Science at Oxford and Cambridge  >>
Learning Techniques for Video Shot Detection
Learning Techniques for Video Shot Detection
Outline
Outline
Introduction
Introduction
What is Shot
What is Shot
Types of Shot-Break
Types of Shot-Break
Shot-Break
Shot-Break
Hard Cut
Hard Cut
Fade
Fade
Dissolve
Dissolve
Wipe
Wipe
Shot Detection Methods
Shot Detection Methods
Shot Detection Methods
Shot Detection Methods
Pervious Approaches to Shot Detection
Pervious Approaches to Shot Detection
Pixel Comparison
Pixel Comparison
Block – Based Approach
Block – Based Approach
Histogram Comparison
Histogram Comparison
Edge Change Ratio
Edge Change Ratio
Comparison…
Comparison…
Problems with previous approaches
Problems with previous approaches
Temporal – Slice Analysis
Temporal – Slice Analysis
Temporal – Slice Analysis
Temporal – Slice Analysis
Cue Video
Cue Video
Temporal – Slice Analysis
Temporal – Slice Analysis
Cue Video
Cue Video
Proposed Approaches
Proposed Approaches
Proposed Approaches
Proposed Approaches
Supervised Learning
Supervised Learning
Feature Extraction
Feature Extraction
How these features can be used to classify images
How these features can be used to classify images
Oops!! There are 15, 625 features
Oops!! There are 15, 625 features
AdaBoost Algorithm
AdaBoost Algorithm
Supervised Learning
Supervised Learning
Unsupervised techniques Clustering
Unsupervised techniques Clustering
Unsupervised technique - clustering
Unsupervised technique - clustering
Unsupervised technique - clustering
Unsupervised technique - clustering
Unsupervised technique
Unsupervised technique
Semi-supervised Learning
Semi-supervised Learning
Semi-supervised Learning
Semi-supervised Learning
Conclusion…
Conclusion…
Conclusion…
Conclusion…
Thank you…
Thank you…

Презентация на тему: «Learning Techniques for Video Shot Detection». Автор: Nithya. Файл: «Learning Techniques for Video Shot Detection.ppt». Размер zip-архива: 938 КБ.

Learning Techniques for Video Shot Detection

содержание презентации «Learning Techniques for Video Shot Detection.ppt»
СлайдТекст
1 Learning Techniques for Video Shot Detection

Learning Techniques for Video Shot Detection

by M. Nithya

Under the guidance of Prof. Sharat Chandran

2 Outline

Outline

Introduction Types of Shot-break Previous approaches to Shot Detection General Approach - pixel comparison, histogram comparison… Recent Work – Temporal slice analysis, Cue Video Our Proposed approaches Supervised Learning using AdaBoost algorithm Unsupervised Learning using clustering Semi-supervised Learning combining AdaBoost & clustering Conclusion

3 Introduction

Introduction

9,000 hours of motion pictures are produced around the world every year. 3,000 television stations broadcasting for twenty-four hours a day produce eight million hours of video per year. Problems: Searching the video Retrieving the relevant information Solution: Break down the video into smaller manageable parts called “Shots”

4 What is Shot

What is Shot

Shot is the result of uninterrupted camera work Shot-break is the transition from one shot to the next

5 Types of Shot-Break

Types of Shot-Break

6 Shot-Break

Shot-Break

Dissolve

Wipe

Fade

Hard Cut

7 Hard Cut

Hard Cut

8 Fade

Fade

9 Dissolve

Dissolve

10 Wipe

Wipe

11 Shot Detection Methods

Shot Detection Methods

12 Shot Detection Methods

Shot Detection Methods

Goal: To segment video into shots Two ways: Cluster the similar frames to identify shots Find the shots that differ and declare it as shot-break

13 Pervious Approaches to Shot Detection

Pervious Approaches to Shot Detection

General Approaches Pixel Comparison Block-based approach Histogram Comparison Edge Change Ratio Recent Work Temporal Slice Analysis Cue Video

14 Pixel Comparison

Pixel Comparison

?x=1 ?y=1 | Pi(x,y) – Pi+1(x,y) |

Frame N + 1

Frame N

D(i,i+1)=

XY

X

Y

15 Block – Based Approach

Block – Based Approach

Frame N + 1

Frame N

Compares statistics of the corresponding blocks

Counts the number of significantly different blocks

16 Histogram Comparison

Histogram Comparison

17 Edge Change Ratio

Edge Change Ratio

18 Comparison…

Comparison…

Method

Advantages

Disadvantages

Pixel-Comparison

Simple, easy to implement

Computationally heavy, Very sensitive to moving object or camera motion

Block based

Performs better than pixel

Can’t identify dissolve, fade, fast moving objects

Histogram comparison

Performance is better Detects hard-cut, fade, wipe and dissolve

Fails if the two successive shots have same histogram. Can’t distinguish fast object or camera motion

Edge Change Ratios

Detects hard-cut, fade, wipe and dissolve

Computationally heavy Fails when there is large amount of motion

19 Problems with previous approaches

Problems with previous approaches

? Can’t distinguish shot-breaks with Fast object motion or Camera motion Fast Illumination changes Reflections from glass, water Flash photography ? Fails to detect long and short gradual transitions

20 Temporal – Slice Analysis

Temporal – Slice Analysis

21 Temporal – Slice Analysis

Temporal – Slice Analysis

22 Cue Video

Cue Video

23 Temporal – Slice Analysis

Temporal – Slice Analysis

24 Cue Video

Cue Video

Graph based approach Each frame maps to a node Connected upto 1, 3 or 7 frames apart Each node is associated with color Histogram Edge Histogram Weights of the edges represent similarity measure between the two frames Graph partitioning will segment the video into shots

25 Proposed Approaches

Proposed Approaches

26 Proposed Approaches

Proposed Approaches

Use learning techniques to distinguish between shot-break and Fast object motion or Camera motion Fast Illumination changes Reflections from glass, water Flash photography

27 Supervised Learning

Supervised Learning

28 Feature Extraction

Feature Extraction

25 Primitive features like edge, color are extracted directly from the image These 25 features are used as input to next round of feature extraction yielding 25 x 25 = 625 features This 625 features can be used as input to compute 625 x 625 = 15, 625 features

29 How these features can be used to classify images

How these features can be used to classify images

30 Oops!! There are 15, 625 features

Oops!! There are 15, 625 features

Applying them to red, green and blue separately will result in 46, 875 features! Can we find few important features that will help to distinguish the images?

Solution : Use AdaBoost to select these features.

31 AdaBoost Algorithm

AdaBoost Algorithm

Input: (x1,y1) (x2,y2) …(xm,ym) where x1,x2,…xm are the images yi = 0,1 for negative and positive examples Let n and p be the number of positive and negative examples Initial weight w1,i = 1/2n if yi= 0 and w1,I = 1/2p if yi = 1 For t= 1,…T: Train one hypothesis hi(x) for each feature and find the error Choose the hypothesis with low error value update the weight: wt+1,i = wt,i * ?t1-et where ei=0,1for xi classified incorrectly or correctly ?t=et/(1-et) Normalize wt+1,I so that it is a distribution Final hypothesis is calculated as

32 Supervised Learning

Supervised Learning

Extract Highly selective features AdaBoost algorithm to select few important features Train the method to detect different shot-breaks

33 Unsupervised techniques Clustering

Unsupervised techniques Clustering

34 Unsupervised technique - clustering

Unsupervised technique - clustering

35 Unsupervised technique - clustering

Unsupervised technique - clustering

Hard Cut

Dissolve

36 Unsupervised technique

Unsupervised technique

Clustering method to cluster into shots Relevance Feedback

37 Semi-supervised Learning

Semi-supervised Learning

38 Semi-supervised Learning

Semi-supervised Learning

?Combination of Supervised and Unsupervised ?Few labeled data are available, using which it works on large unlabeled video Steps: AdaBoost algorithm to select features Clustering method to cluster into shots Relevance Feedback

39 Conclusion…

Conclusion…

40 Conclusion…

Conclusion…

Problems with previous approaches: Can’t distinguish shot-breaks with Fast object motion or Camera motion Fast Illumination changes Reflections from glass, water Flash photography Fails to detect long and short gradual transitions Planning to use AdaBoost learning based clustering scheme for shot-detection

41 Thank you…

Thank you…

«Learning Techniques for Video Shot Detection»
http://900igr.net/prezentacija/anglijskij-jazyk/learning-techniques-for-video-shot-detection-226396.html
cсылка на страницу

Курсы английского

25 презентаций о курсах английского
Урок

Английский язык

29 тем
Слайды