multimodal analysis for bridging semantic gap with biologically inspired algorithm

78
Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithms Dr. Krishna Chandramouli Media Engineering and Analytics Research Group VIT University

Upload: techkrish

Post on 21-Aug-2015

79 views

Category:

Engineering


0 download

TRANSCRIPT

Page 1: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithms

Dr. Krishna ChandramouliMedia Engineering and Analytics Research Group

VIT University

Page 2: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Overview Who we are!! Media and Internet Information Access Subjective vs Objective Indexing The Semantic Gap Evolving Strategies Social Media Analysis MediaEval 2013 Participation Conclusion Q & A

04/07/2014Uni. of Siegen

Page 3: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Who we are!!04/07/2014Uni. of Siegen

Page 4: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Who we are!!04/07/2014Uni. of Siegen

Page 5: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Media and internet

04/07/2014Uni. of Siegen

Page 6: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Media and internet In March 2013 that Flickr had a

total of 87 million registered members and more than 3.5 million new images uploaded daily.

There are currently almost 90 billion photos total on Facebook.  This means we are, by far, the largest photos site on the Internet.

04/07/2014Uni. of Siegen

Page 7: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Information AccessTextual searchVisual searchSearch query formulation

04/07/2014Uni. of Siegen

Page 8: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Information Access Traditional ordering of images is achieved through

categorization of information into logical structures Creation of albums Categorizing through date/time Clustering through location

Image based search engines are gaining popularity with the increase in power of indexing schemes

04/07/2014Uni. of Siegen

Page 9: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Information Access04/07/2014Uni. of Siegen

Page 10: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Information Access04/07/2014Uni. of Siegen

Page 11: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Information Access04/07/2014Uni. of Siegen

Page 12: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Information Access04/07/2014Uni. of Siegen

Page 13: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Indexing subjective or objective

04/07/2014Uni. of Siegen

Page 14: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Indexing subjective or objective How to uniquely name an image to make them distinguishable?

What names can be used to search images?

How many names are needed to make the images unique?

Will all humans use the same names to identify the images?

04/07/2014Uni. of Siegen

Page 15: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Indexing subjective or objective Humans are culturally influenced

Terms contain different meanings across boundaries and cultures

Therefore, any tag/word assigned to an image will be considered subjective

Objective signatures for images are generated from the characteristics of the images

The beginning of MPEG-7 standardisation activities.

04/07/2014Uni. of Siegen

Page 16: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Indexing subjective or objective Image characteristics exploited for objective annotation include

Colour Colour Layout Descriptor Colour Structure Descriptor Dominant Colour Descriptor Scalable Colour Descriptor

Texture Texture Browsing Descriptor Edge Histogram Descriptor Homogenous Texture Descriptor

Shape

04/07/2014Uni. of Siegen

Page 17: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

The Semantic Gap

04/07/2014Uni. of Siegen

Page 18: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

The Semantic Gap The semantic gap characterizes the difference between two

descriptions of an object by different linguistic representations, for instance languages or symbols.

In computer science, the concept is relevant whenever ordinary human activities, observations, and tasks are transferred into a computational representation

04/07/2014Uni. of Siegen

Page 19: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

The Semantic Gap04/07/2014Uni. of Siegen

Page 20: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

The Semantic Gap04/07/2014Uni. of Siegen

Page 21: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Evolving strategiesImage Classification; Visual Classifier; Knowledge Assisted Analysis; Image Retrieval and User Relevance Feedback; Multi-Concept Space Search and Retrieval

04/07/2014Uni. of Siegen

Page 22: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Evolving strategies The problem of Image classification and clustering has been

the subject of active research for last decade. Mainly attributed to the exponential growth of digital content.

The efficiency of the clustering and classification algorithms can be attributed to the efficiency of the machine learning approaches.

To improve the performance of machine learning algorithms, different optimisation techniques has been employed such as Genetic Algorithms.

04/07/2014Uni. of Siegen

Page 23: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Evolving strategies Recent developments in applied and heuristic optimisation

techniques have been strongly influenced and inspired by natural and biological systems.

Algorithms developed from such observations are Ant Colony Optimisation (ACO) - based on the ability of an ant colony to

nd the shortest path between the food and the source compared to an individual ant.

Articial Immune System (AIS) - typically exploit the immune system's characteristics of learning and memory to solve a problem

Particle Swarm Optimisation (PSO) - inspired by the social behaviour of a flock of birds.

04/07/2014Uni. of Siegen

Page 24: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Evolving strategies In the study of "Semantic Gap", machine learning algorithms

are the building blocks for bottom-up approach.

Some of the applications of efficient machine learning algorithms are: Automatic Content Annotation Knowledge Extraction Content Retrieval

In the top-down approach, Ontology provides partial understanding of human semantics.

04/07/2014Uni. of Siegen

Page 25: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier

04/07/2014Uni. of Siegen

Page 26: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier In an effort to transform the social interaction of different species into a

computer simulation, Kennedy and Eberhart developed an optimisation technique named Particle Swarm Optimisation.

In theory, the universal behaviour of individuals is summarised in terms of Evaluate, Compare and Imitate principles.

04/07/2014Uni. of Siegen

Page 27: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier Evaluate: The tendency to evaluate stimuli – to rate them as

positive or negative, attractive or repulsive is perhaps the most ubiquitous behavioural characteristic of living organisms.

Compare: In almost every aspect of life, human tend to compare with others

Imitate: Humans imitation comprises taking the perspective of the other person, not only imitating a behaviour but also realising its purpose and executing the behaviour when it is appropriate

04/07/2014Uni. of Siegen

Page 28: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier Equations governing the motion of particles in PSO.

04/07/2014Uni. of Siegen

Page 29: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier Pseudo code for the algorithm

Step 1: Random Initialization of Particles Step 2: Function Evaluation Step 3: Computation of personal best and global best Step 4: Velocity update Step 5: Position update Step 6: Loop to step 2, until the stopping criteria is reached

04/07/2014Uni. of Siegen

Page 30: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier Self Organising Map

04/07/2014Uni. of Siegen

[X]

[X] - Input feature vectorClass 1 – RedUntrained - Black

Winner Node selected based on L2 norm

Page 31: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier04/07/2014Uni. of Siegen

Page 32: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier04/07/2014Uni. of Siegen

.. .

Winner Node

)]([)()1( tmxhtmtm iciii )]([)()1( tmxhtmtm iciii

Dual Layer SOM

Page 33: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Visual classifier The elementary principle of “Chaos” is introduced to model the behaviour

of particle motion. The theoretical discussion on Chaotic – PSO includes the notion of “wind

speed” and “wind direction” modelling the biological atmosphere for position update of the particles.

The wind speed and therefore the position update equation are presented by:

04/07/2014Uni. of Siegen

Page 34: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework

04/07/2014Uni. of Siegen

Page 35: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework

04/07/2014Uni. of Siegen

Page 36: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework

04/07/2014Uni. of Siegen

Page 37: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework Experimental Dataset

A set of 500 Images, belonging to the general category of vacation images was assembled.

The content was mainly obtained from Flickr online photo management and sharing application and includes images that depict cityscape, seaside, mountain and landscape locations.

Every image was manually annotated, i.e. after the segmentation algorithm is applied, a single concept was associated with each resulting image segment

04/07/2014Uni. of Siegen

Page 38: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework

04/07/2014Uni. of Siegen

Page 39: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework

04/07/2014Uni. of Siegen

Page 40: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Knowledge Assisted Framework From the results it can be seen that the combined use of PSO

optimisation technique with SOM results in better classification accuracy compared to using the latter alone.

It can be noted that the performance of PSO classier is better than the performance of SVM and GA classifiers.

Since, SVM's need large training data to accurately discriminate between image classes.

04/07/2014Uni. of Siegen

Page 41: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Image Retrieval and User Relevance Feedback

04/07/2014Uni. of Siegen

Page 42: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

User Relevance Feedback04/07/2014Uni. of Siegen

Page 43: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

User Relevance Feedback04/07/2014Uni. of Siegen

Page 44: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

User Relevance Feedback The database used in the experiment is generated from Corel

Dataset and consists of seven concepts namely, building, cloud, car, elephant, grass, lion and tiger

The test set has been modelled for seven concepts with a variety of background elements and overlapping concepts, hence making the test set complex.

04/07/2014Uni. of Siegen

Page 45: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

User Relevance Feedback04/07/2014Uni. of Siegen

Page 46: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

User Relevance Feedback04/07/2014Uni. of Siegen

Page 47: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space

04/07/2014Uni. of Siegen

Page 48: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space04/07/2014Uni. of Siegen

• High-level queries“A tiger resting in the

forest and guarding his territory”

• Mid-level features (context independent)

“Tiger”, “Grass”, “Rock”, “Water”,……

Page 49: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space• Mid-level features:

In a constrained environment with limited number of mid-level features, the performance of classification algorithm has found to be satisfactory

• High-level queries: Open to subjective interpretation of the concepts and also may involve

more than one mid-level feature

Main objective:• In this multi-concept framework, users are encouraged to construct

high level queries based on their preferences

04/07/2014Uni. of Siegen

Page 50: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space04/07/2014Uni. of Siegen

Page 51: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space• SVM Classifier

• SVM Light toolbox was used to generate semantic labels• CLD+EHD

• Multi-feature classifier (MF) • Employs a mixture of 7 visual features.

• The visual features are merged using Multi-Objective Learning (MOL)

04/07/2014Uni. of Siegen

Page 52: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space Pre-processing stage: mid-level feature concept detection

Query formulation: users to construct a high-level semantic information space

04/07/2014Uni. of Siegen

Page 53: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space Fisheye distortion

technique

Overview + focus

04/07/2014Uni. of Siegen

Page 54: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space• Query space panel

• Concept map panel

• Concept chart panel

04/07/2014Uni. of Siegen

Page 55: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space A 3500 image set collection

From Corel dataset Natural images with many elements Foreground and background Rich semantic context Fully annotated

10 mid-level concepts lion, water, grass, building, car, cloud, rock, tiger, elephant, flower

8 high-level concepts flower fields, modern city view, rural garden, mountain view, waterfalls, wild life,

city street, boat

04/07/2014Uni. of Siegen

Page 56: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space Retrieval of high level queries using the proposed MCB framework

04/07/2014Uni. of Siegen

Page 57: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space Retrieval of high level queries using SVM classification

04/07/2014Uni. of Siegen

Page 58: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space Content-based retrieval with RF mechanism

04/07/2014Uni. of Siegen

Page 59: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space04/07/2014Uni. of Siegen

Page 60: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space04/07/2014Uni. of Siegen

Page 61: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Multi-concept search space04/07/2014Uni. of Siegen

Landscape water, grass 0.58

Modern city building, cloud 0.8Wild life lion, tiger, elephant 0.59Rural garden flower, water, grass 0.9

User 2Landscape water 0.23

Modern city building 0.71Wild life lion, rock, grass, tiger, elephant 0.87Rural garden flower 0.28

User 3Landscape water, grass, cloud, car, elephant 0.59Modern city cloud, building, car 0.91

Wild life lion, tiger, grass, elephant, rock 0.82Rural garden flower, water, grass 0.88

Page 62: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis

04/07/2014Uni. of Siegen

Page 63: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis Social media is the interaction among people in which they create, share

or exchange information and ideas in virtual communities and networks.

Andreas Kaplan and Michael Haenlein define social media as "a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0

Social media allows for the creation and exchange of user-generated content.

Social media differ from traditional or industrial media in many ways, including quality, reach, frequency, usability, immediacy, and permanence.

04/07/2014Uni. of Siegen

Page 64: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis• Images are often accompanies with free-text annotations,

which can be used as complementary information for content-based classification

• The challenge is to extract entities from text and classify them into an arbitrary set of classes

04/07/2014Uni. of Siegen

Plansarsko lakeShepherd in Bucegi National Park

Page 65: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis04/07/2014Uni. of Siegen

Page 66: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis04/07/2014Uni. of Siegen

Page 67: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis04/07/2014Uni. of Siegen

Page 68: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis04/07/2014Uni. of Siegen

Page 69: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Social Media Analysis Content-based analysis (KAA)

restricted to classes for which the classifier has been learnt

For text-based analysis (SCM/THD), the classes have to be exhaustive - all entities are classified

Mapping from SCM/THD to KAA

Perform intersection between the individual classifier results

Select concept occupying largest area on the image

04/07/2014Uni. of Siegen

Page 70: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

MediaEval 2013 Participation

04/07/2014Uni. of Siegen

Page 71: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

VIT @ MediaEval 2013 Social Event Detection Task

18/04/23

Page 72: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

VIT @ MediaEval 201318/04/23

The geographical coordinates is an important component and indicator of where an event has happened.

The event clusters are analysed through the weighted occurrence of tags among the distribution of media annotation

Page 73: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

VIT @ MediaEval 201318/04/23

The system computes the similarity between synset representing the tags and each of the categories.

We use Lin similarity measure to evaluate the semantic distance between the synset and category.

Page 74: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

VIT @ MediaEval 2013 Placing Task

18/04/23

Page 75: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

VIT @ MediaEval 2013 Dividing the globe into grids with a maximum of 10,000

images per grid . Starting from an initial grid that spans the entire globe, recursively subdividing grids into smaller ones once the threshold is reached.

18/04/23

Page 76: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Conclusion and Future Work

04/07/2014Uni. of Siegen

Page 77: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Conclusion Automatic concept detection within images is a challenging and as of yet

unsolved research problem.

Impressive improvements have been achieved, although most of the proposed systems rely on training data that has been manually, and thus reliably labeled, an expensive and laborious endeavor that cannot easily scale.

Current research in domain adaptation focuses on a scenario where (a) the prior domain (source) consists of one or maximum two databases (b) the labels between the source and the target domain are the same, and (c) the number of annotated training data for the target domain are limited.

04/07/2014Uni. of Siegen

Page 78: Multimodal Analysis for Bridging Semantic Gap with Biologically Inspired Algorithm

Thank you for your attention

Q & A