MultiMedia Modeling

Name: MultiMedia Modeling | 22nd International Conference, MMM 2016, Miami, FL, USA, January 4-6, 2016, Proceedings, Part I
Brand: Springer
Price: 96.29 EUR
Availability: OnlineOnly

22nd International Conference, MMM 2016, Miami, FL, USA, January 4-6, 2016, Proceedings, Part I

Qi Tian Nicu Sebe Guo-Jun Qi Benoit Huet Richang Hong Xueliang Liu(Herausgeber*in)

Springer (Verlag)

Erschienen am 2. Januar 2016

XXIV, 927 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-319-27671-7 (ISBN)

96,29 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents - Part I
Contents - Part II
Regular Papers
Video Event Detection Using Kernel Support Vector Machine with Isotropic Gaussian Sample Uncertainty (KSVM-iGSU)
1 Introduction
2 Related Work
3 Kernel SVM-iGSU
3.1 Overview of LSVM-iGSU
3.2 Kernelizing LSVM-iGSU (KSVM-iGSU)
3.3 Relevance Degree KSVM-iGSU
4 Experiments and Results
4.1 Dataset and Evaluation Measures
4.2 Video Representation and Uncertainty
4.3 Experimental Results and Discussion
5 Conclusions and Future Work
References
Video Content Representation Using Recurring Regions Detection
1 Introduction
2 Related Work
3 Recurring Region Detection
3.1 Region Detection and Representation
3.2 Tracking, Shot-Track Representation, and Merging
3.3 Shot-Track Matching and Video-Track Merging
4 Evaluation
4.1 Dataset
4.2 Quantitative Evaluation
4.3 Qualitative Evaluation
5 Conclusion
References
Group Feature Selection for Audio-Based Video Genre Classification
1 Introduction
2 Feature Selection
3 Evaluation Setup
3.1 Audio Features
3.2 Datasets
3.3 Data Preprocessing
3.4 Classification
3.5 Performance Metrics
4 Experimental Results
4.1 Classification Performance
4.2 Feature Analysis
5 Conclusion
References
Computational Cartoonist: A Comic-Style Video Summarization System for Anime Films
1 Introduction
2 Related Work
3 Comic-Style Video Summarization
3.1 Shot Transition Detection
3.2 Key Frame Detection
3.3 Comic Layout
4 Evaluation
5 Conclusions and Future Work
References
Exploring the Long Tail of Social Media Tags
1 Introduction
2 Related Work
3 What Tags Constitute the Long Tail?
4 Utilizing the Long Tail
4.1 Augmenting Rare Tags
4.2 Tag Relevance
4.3 Learning Detectors
5 Conclusions
References
Visual Analyses of Music Download History: User Studies
1 Introduction
2 Related Work
3 Visualization Design
4 User Studies
4.1 Implementation
4.2 Participants
4.3 Tasks
4.4 Results
4.5 Discussion
5 Conclusions
References
Personalized Annotation for Mobile Photos Based on User's Social Circle
1 Introduction
2 Related Work
3 Proposed Framework
3.1 Label Generation
3.2 Label Propagation
4 Experiment
4.1 DataSet
4.2 Evaluation of Album
4.3 Evaluation of Tag Generation
4.4 Evaluation of Personalized Annotation
5 Conclusion
References
Utilizing Sensor-Social Cues to Localize Objects-of-Interest in Outdoor UGVs
1 Introduction
2 Related Work
3 Sensor-Social-Based OOI Recognition
3.1 OOI Acquisition from UGVs
3.2 Classified Object Set Recommendation
3.3 OOI Description and Recognition
4 Experimental Results and Analysis
4.1 Dataset and Experimental Setup
4.2 Experimental Results and Analysis
5 Conclusions
References
NEWSMAN: Uploading Videos over Adaptive Middleboxes to News Servers in Weak Network Infrastructures
1 Introduction
2 Related Work
3 System Overview
4 Scheduling Problem and Solution
5 Simulations and Results
6 Conclusions
References
Computational Face Reader
1 Introduction
2 Related Work
2.1 Face Reading
2.2 Deep Convolutional Neural Network
3 Overview of Our Framework
4 Dataset Preparation and Library Construction
5 Deep Networks with Facial Region Pooling
5.1 Architecture
5.2 Training the Network
6 Experiments
6.1 Evaluation of Facial Attribute Estimation
6.2 Evaluation of Face Reading
7 Conclusions and Future Work
References
Posed and Spontaneous Expression Recognition Through Restricted Boltzmann Machine
1 Introduction
2 The Proposed Method
2.1 Feature and Facial Events Extraction
2.2 Posed and Spontaneous Expression Modeling Using RBM
3 Experiments and Analysis
3.1 Experimental Conditions
3.2 Experimental Results and Analysis
3.3 Comparison with Other Methods
4 Conclusion and Future Works
References
DFRS: A Large-Scale Distributed Fingerprint Recognition System Based on Redis
1 Introduction
2 Background
2.1 Redis
2.2 Fingerprint Feature
2.3 Fingerprint Recognition
3 DFRS
3.1 System Architecture
3.2 Encoding and Decoding
3.3 Process of Fingerprint Recognition
3.4 Match Strategy Based on Quick-Return
4 Evaluation
4.1 Experiment Environment
4.2 Performance of Encoding and Decoding
4.3 Analytical Evaluation for Matching Method
4.4 Evaluation for Quick-Return
5 Conclusion and Future Work
References
Logo Recognition via Improved Topological Constraint
1 Introduction
2 Improved Topological Constraint
3 Logo Recognition
3.1 Feature Selection
3.2 Recognition
4 Experiments
4.1 Impact of Parameters
4.2 FlickrLogos-32 Dataset
4.3 FlickrLogos-27 Dataset
5 Conclusion
References
Compound Figure Separation Combining Edge and Band Separator Detection
1 Introduction
2 Related Work and Context
3 Proposed Algorithm
3.1 Illustration Classifier
3.2 Recursive Algorithm
3.3 Edge-Based Separator Detection
3.4 Band-Based Separator Detection
4 Parameter Optimization
5 Evaluation
5.1 Evaluation on ImageCLEF Dataset
5.2 Evaluation on NLM Dataset
5.3 Illustration Classifier Accuracy
6 Conclusion and Further Work
References
Camera Network Based Person Re-identification by Leveraging Spatial-Temporal Constraint and Multiple Cameras Relations
1 Introduction
2 Observations
3 Our Approach
3.1 Problem Definition
3.2 Probabilistic Model with Spatial-Temporal Constraint
3.3 Optimization with Multiple Camera Relations
4 Experiments
4.1 Baselines
4.2 TMin Data Set
4.3 CamNeT
4.4 Running Time
5 Conclusion
References
Global Contrast Based Salient Region Boundary Sampling for Action Recognition
1 Introduction
2 Methodology
2.1 Improved Dense Trajectories
2.2 Motion Boundary Based Sampling
3 Our Approach
3.1 Global Contrast Based Salient Region Sampling
3.2 Optimization with Salient Region Boundary
4 Experiments
4.1 Datasets
4.2 Experimental Setup
4.3 Results and Analysis
5 Conclusion
References
Elastic Edge Boxes for Object Proposal on RGB-D Images
1 Introduction
2 Related Work
3 Elastic Edge Boxes
3.1 Initial Bounding Boxes Generation
3.2 Elastic Range Determination
3.3 Bounding Box Adjustment
4 Experiments
4.1 Dataset Construction
4.2 Performance Evaluation
5 Conclusions
References
Pairing Contour Fragments for Object Recognition
Abstract
1 Introduction
2 Related Works
3 Contour Fragment Pairs
3.1 Matching Energy
3.2 Pairing Algorithm
3.3 Learning Subspace for CFPs
4 Recognition Algorithm
4.1 Learning Weak Classifiers Based on CFPs
4.2 Voting Boundaries and Foregrounds
5 Experiments
5.1 Experiments on Weizmann Horses Dataset
5.2 Experiments on ETHZ Shape Dataset
6 Conclusion and Future Works
References
Instance Search with Weak Geometric Correlation Consistency
1 Introduction
2 Related Work
3 Weak Geometric Correlation Consistency
3.1 Motivation
3.2 Implementation
3.3 Computational Complexity
4 Experiments
4.1 Datasets
4.2 Evaluation Protocol
4.3 Experiment Settings
5 Results and Discussion
6 Conclusion
References
Videopedia: Lecture Video Recommendation for Educational Blogs Using Topic Modeling
1 Introduction
2 Related Work
3 System Model
3.1 Dataset Used for Recommendation
3.2 Extracting Video Content
3.3 Extracting Webpage Content
3.4 Processing the Extracted Text
3.5 Topic Modeling
3.6 Definition of Similarity Matching
3.7 Algorithm for Video Recommendation
4 Experimental Results
4.1 Baselines Used for Comparison
4.2 Evaluation of Recommendation
5 Conclusions
References
Towards Training-Free Refinement for Semantic Indexing of Visual Media
1 Introduction
2 Related Work
3 Motivation and Proposed Solution
4 Training-Free Refinement (TFR)
4.1 Factorizing Detection Results
4.2 Integration with Ontologies
4.3 Temporal Neighbourhood-Based Propagation
5 Experiments and Discussion
5.1 Evaluation on Wearable Camera Images (Dataset1)
5.2 Evaluation on TRECVid Video (Dataset2)
5.3 Efficiency Analysis of TFR
6 Conclusions
References
Deep Learning Generic Features for Cross-Media Retrieval
1 Introduction
2 Related Work
3 Deep Learning Generic Features
3.1 Layer-Wise Pre-training
3.2 Overall Fine-Tuning
3.3 Cross-Media Retrieval
4 Experiments and Results
4.1 Experimental Setup
4.2 Experimental Results
5 Conclusions
References
Cross-Media Retrieval via Semantic Entity Projection
1 Introduction
2 Related Works
3 Our Approach
3.1 Entity Level Construction
3.2 Entity Projection Learning
3.3 Semantic Abstraction Generation
4 Experimental Evaluation
4.1 Dataset Description
4.2 Evaluation Metrics
4.3 Experimental Results
5 Conclusions
References
Visual Re-ranking Through Greedy Selection and Rank Fusion
1 Introduction
2 Effective Visual Re-ranking
2.1 Informative Feature Extraction
2.2 Label De-noising
2.3 Graph-Based Re-ranking
2.4 Multiple Graph Fusion
3 Experiments
3.1 Dataset
3.2 Performance Metrics
3.3 Experiments for Label De-noising
3.4 Experiments for Greedy Selection
3.5 Experiments for Rank Fusion
3.6 Experiments to Compare with the State-of-Art Methods
4 Conclusion
References
No-reference Image Quality Assessment Based on Structural and Luminance Information
1 Introduction
2 The Proposed NR IQA Model
2.1 Spatial Divisive Normalization
2.2 The LBP Histogram
2.3 The Normalized Luminance Histogram
2.4 Regression Model for Quality Prediction
3 Experimental Results and Analysis
3.1 Implementation Details
3.2 Database Description and Evaluation Methodology
3.3 Experimental Results
4 Conclusions
References
Learning Multiple Views with Orthogonal Denoising Autoencoders
1 Introduction
2 Related Work
3 Approach
3.1 Problem Formulation
3.2 Basic Autoencoder
3.3 Orthogonal Autoencoder for Multi-view Learning
3.4 Training of Orthogonal Autoencoder
3.5 Orthogonal Denoising Autoencoder for Robust Latent Spaces
4 Experiments
4.1 Synthetic Dataset
4.2 Real-World Dataset
5 Conclusions
References
Fast Nearest Neighbor Search in the Hamming Space
1 Introduction
2 Related Works
2.1 Multi-index Hashing
2.2 FLANN
3 Our Approach
3.1 Data Structure
3.2 Search over the Augmented Neighborhood Graph
4 Experiments
4.1 Datasets and Settings
4.2 Results
4.3 Analysis
5 Conclusions
References
SOMH: A Self-Organizing Map Based Topology Preserving Hashing Method
1 Introduction
2 Background
2.1 Vector Quantization
2.2 Self-Organizing Map
3 SOMH
3.1 Naive SOMH
3.2 Relaxed SOMH
3.3 An Iterative Solution
3.4 Product Space SOMH
4 Experiments
4.1 Dataset
4.2 Baselines
4.3 Performance Evaluation on Short Binary Code
4.4 Performance Evaluation on Long Hashing Code
4.5 Training Time Evaluation
4.6 Parameter Evaluation
5 Conclusion
References
Describing Images with Ontology-Aware Dictionary Learning
1 Introduction
2 Ontology-Aware Dictionary Learning
3 Solution and Algorithm
4 Experiments
4.1 Datasets and Parameters
4.2 Comparison Methods
4.3 Evaluation Metric
4.4 Experimental Results
5 Conclusions
References
Quality Analysis on Mobile Devices for Real-Time Feedback
1 Introduction
1.1 Application Scenario
1.2 Related Work
2 Quality Analysis Algorithms
2.1 Sharpness
2.2 Noise
2.3 Over-/Underexposure
3 Implementation
4 Evaluation
4.1 Sharpness
4.2 Noise
5 Conclusion
References
Interactive Search in Video: Navigation With Flick Gestures vs. Seeker-Bars
1 Introduction
2 Related Work
3 Video Navigation with Flick Gestures
3.1 Interaction Concept
3.2 Implementation Details and Issues
4 Evaluation
4.1 Target Search Tasks
4.2 Scene Counting Tasks
4.3 Questionnaires
4.4 Preferred Interface
5 Discussion and Conclusions
References
Second-Layer Navigation in Mobile Hypervideo for Medical Training
1 Introduction
1.1 Problem Statement
1.2 Research Contributions
2 Related Work
3 Usability Evaluation in the Design Phase
3.1 Expert Group
3.2 Survey
3.3 User Test with Prototype
4 Implementation
5 Final Evaluation
6 Conclusion
References
Poster Papers
Reverse Testing Image Set Model Based Multi-view Human Action Recognition
Abstract
1 Introduction
2 Reverse Testing Image Set Model Based Multi-view Human Action Recognition
2.1 The Scheme of Adding Samples in Query Set
2.2 Reverse Testing Image Set Model Based Multi-view Action Recognition Model
2.3 Solution and Inference
3 Experimental and Discussion
3.1 Experimental Setting
3.2 Evaluation the Relationships of Different Views
3.3 Evaluation the Effect of the Number of Samples in Query Set - RTIS
3.4 Performance Evaluation of the Proposed Algorithm
4 Conclusions
Acknowledgments
References
Face Image Super-Resolution Through Improved Neighbor Embedding
1 Introduction
2 Notations
3 Position-Patch Based Face Image Super-Resolution
3.1 Least Square Representation
3.2 Sparse Representation
3.3 Locality-Constrained Representation
4 Face Image Super-Resolution Through Tikhonov Regularized Neighbor Representation (TRNR)
5 Experiments and Result Analysis
6 Conclusion
References
Adaptive Multichannel Reduction Using Convex Polyhedral Loudspeaker Array
1 Introduction
2 Related Work
2.1 Sound Fields Reproduction Model
2.2 The Conversion Method
3 The Proposed Reduction Method
3.1 The Reduction Scheme
3.2 Convex Polyhedral Loudspeaker Array
3.3 Error Metric
4 Simulation and Subjective Evaluation Results
4.1 Simulation
4.2 Example Loudspeaker Arrays and Sound Fields
4.3 Conversion Error
4.4 Subjective Evaluation
5 Conclusion
References
Dominant Set Based Data Clustering and Image Segmentation
1 Introduction
2 Dominant Sets Clustering
3 Our Algorithm
4 Experiments
4.1 Data Clustering
4.2 Image Segmentation
5 Conclusions
References
An R-CNN Based Method to Localize Speech Balloons in Comics
Abstract
1 Introduction
2 Introduction to R-CNN
2.1 Detection Process
2.2 Training Process
3 Experiment
3.1 Dataset
3.2 Performance Evaluation Criteria
3.3 Experimental Result
4 Conclusion
References
Facial Age Estimation with Images in the Wild
1 Introduction
2 Related Work
3 Building an Aging Collection in the Wild
3.1 Data Collection
3.2 Face Detection and Alignment
4 Cost-Sensitive Learning for Age Estimation
4.1 Biased Penalties SVM
4.2 Random Forests
4.3 Cost Function
5 Experiments
5.1 Dataset and Feature Extraction
5.2 Within-Database Experiments
5.3 Cross-Database Experiments
6 Conclusion
References
Fast Visual Vocabulary Construction for Image Retrieval Using Skewed-Split k-d Trees
1 Introduction
2 Related Work
3 The Exponential Distribution of SIFT Descriptors
4 Visual Vocabulary Construction Using k-d Trees with Skewed Split
4.1 Construction of k-d Trees with Skewed Split
4.2 Clustering Using a Forest of 8 k-d Trees with Skewed Split
5 Application to Image Retrieval
6 Conclusion
References
OGB: A Distinctive and Efficient Feature for Mobile Augmented Reality
Abstract
1 Introduction
2 State-of-the-Art Binary Descriptors
3 ORB: Oriented Gradient Binary
3.1 OGB Extraction Process
3.2 Comparison with Lightweight Binaries
3.3 Selection of Parameters and Settings
4 Applications on Mobile Devices
4.1 Mobile Object Recognition
4.2 Real-Time Mobile Object Tracking
5 Conclusion
References
Learning Relative Aesthetic Quality with a Pairwise Approach
1 Introduction
2 Related Work
3 Relative Aesthetic Quality Ranking
3.1 Pairwise-Based Ranking Model
3.2 Training Pairs Generation
3.3 Informative Training Pairs Selection
4 Experiments
4.1 Datasets
4.2 Experimental Settings
4.3 Experimental Results on CUHKPQ
4.4 Experimental Results on AVA
5 Conclusion and Future Work
References
Robust Crowd Segmentation and Counting in Indoor Scenes
Abstract
1 Introduction
2 Related Work
3 Robust Crowd Counting
3.1 Pre-processing
3.2 Crowd Segmentation
3.3 Crowd Normalization and Counting
4 Experimental Results
5 Conclusion
References
Robust Sketch-Based Image Retrieval by Saliency Detection
1 Introduction
2 Related Work
3 Algorithm
3.1 Saliency Detection
3.2 Gradient Field
3.3 Multi-scale HOG
3.4 Sketch-Based Image Retrieval
4 Results and Discussions
5 Conclusion
References
Image Classification Using Spatial Difference Descriptor Under Spatial Pyramid Matching Framework
Abstract
1 Introduction
2 Related Work
3 Proposed Framework
3.1 Feature Extraction
3.2 Sparse Coding
3.3 Spatial Pooling
3.4 Spatial Difference Descriptor Computation
4 Experiments and Results
4.1 Scene 15 Dataset
4.2 Caltech 101 Dataset
4.3 Caltech 256 Dataset
5 Conclusion
References
Exploring Relationship Between Face and Trustworthy Impression Using Mid-level Facial Features
Abstract
1 Introduction
2 Related Works
3 Proposed Method
3.1 Low-level Features
3.2 Mid-level Features
4 Experiments
4.1 Dataset
4.2 Experiments and Discussions
5 Conclusions
References
Edit-Based Font Search
1 Introduction
2 Related Work
2.1 Generating, Learning, and Searching Fonts
2.2 Sketch-Based Retrieval Method
3 Edit-Based Font Search Method
3.1 Framework of Edit-Based Font Search
3.2 Requirements for Font Search
4 Font Search Application
5 Experimental Evaluation
5.1 Details of Experiment
5.2 Experimental Results
6 Conclusion
References
Private Video Foreground Extraction Through Chaotic Mapping Based Encryption in the Cloud
1 Introduction
2 Cryptography Primitive
3 Private Video Foreground Extraction
3.1 Random Inverse
3.2 Frame Confusion
3.3 Frame Diffusion
3.4 Foreground Extraction
4 Experimental Results
4.1 The Correctness Rate
4.2 The Extraction Results
4.3 Security Analysis
5 Conclusion and Discussion
References
Evaluating Access Mechanisms for Multimodal Representations of Lifelogs
1 Introduction
2 Background to Lifelogging and Pervasive Access
3 An End-to-End Holistic Lifelogging Solution
4 Modeling for Multimodal Access
5 User Evaluation
5.1 Experimental Dataset
5.2 The Interface Evaluation Process
5.3 Results of the Interface Evaluation
6 Discussion
References
Analysis and Comparison of Inter-Channel Level Difference and Interaural Level Difference
1 Introduction
2 Compare ICLD with ILD Theoretically
2.1 Calculation of ICLD
2.2 Generation and Estimation of ILD
3 Experiments
3.1 Experimental Data Generation
3.2 Experimental Comparison
4 Conclusion
References
Automatic Scribble Simulation for Interactive Image Segmentation Evaluation
1 Introduction
2 Related Work
3 Data Set
4 Analysis of Scribble Variety
4.1 Scribble Difference
4.2 Influence on Segmentation Result
5 Automatic Scribble Simulation
5.1 Scribble Consistency on Superpixel and Superpixel Group Levels
5.2 Distribution of Superpixel Group Coverage
5.3 Distribution of Superpixel Coverage
5.4 Effect of Connection in Scribble
5.5 Scribble Simulation
6 Evaluation of Interactive Segmentation Algorithms
7 Conclusion
References
Multi-modal Image Re-ranking with Autoencoders and Click Semantics
1 Introduction
2 Multimodal Learning with Autoencoders and Click Semantics
2.1 Marginalized Denoising Autoencoders
2.2 Manifold Learning with Click Semantics
3 Experimental Evaluation
3.1 Experiment Introduction
3.2 Experimental Results
4 Conclusion
References
Sketch-Based Image Retrieval with a Novel BoVW Representation
Abstract
1 Introduction
2 Visual Vocabulary Generation
3 Sketch-Image Matching via Quantization
4 Indexing Structure Construction
5 Experiment and Analysis
6 Conclusions and Future Work
References
Symmetry-Aware Human Shape Correspondence Using Skeleton
1 Introduction
2 Related Work
3 Skeleton-Based Symmetry-Aware Approach
3.1 Correspondence Candidate Set
3.2 Skeleton-Based Symmetry-Aware Shape Correspondence
3.3 One-to-One Correspondence
4 Experiments
4.1 Dataset
4.2 Performance Evaluation
5 Conclusion
References
XTemplate 4.0: Providing Adaptive Layouts and Nested Templates for Hypermedia Documents
1 Introduction
2 Adaptive Layouts and Nested Templates
2.1 Adaptive Layouts
2.2 Template Nesting
3 Related Work
4 XTemplate 4.0
4.1 Template Processing
4.2 Template Example
5 XTemplate 4.0 Evaluation
6 Conclusions
References
Level Ratio Based Inter and Intra Channel Prediction with Application to Stereo Audio Frame Loss Con ...
Abstract
1 Introduction
2 Level Ratio Based Inter and Intra Channel Prediction
2.1 Inter-channel Prediction
2.2 Intra-channel Prediction
2.3 Level Ratio Based Inter and Intra Channel Prediction
3 Experiments
3.1 Objective Evaluation
3.2 Subjective Evaluation
4 Conclusion and Future Work
References
Depth Map Coding by Modeling the Locality and Local Correlation of View Synthesis Distortion in 3-D Video
Abstract
1 Introduction
2 Proposed VSD Model
2.1 Previous Work on View Synthesis Distortion
2.2 Locality of VSD
2.3 VSD Model Based on Locality and Local Correlation
3 Rate-Distortion Optimization Using the VSD Model
4 Experimental Results
5 Conclusion
Acknowledgement
References
Discriminative Feature Learning with an Optimal Pattern Model for Image Classification
Abstract
1 Introduction
2 The Proposed Framework
2.1 From Images to the Database of Transactions
2.2 Pattern Mining Based on MDL Principle
2.3 Term Relevance Ratio Weighting
3 Experiments
3.1 Datasets and Experimental Setups
3.2 Results and Discuss
4 Conclusions
Acknowledgements
References
Sign Language Recognition Based on Trajectory Modeling with HMMs
1 Introduction
2 System Overview
3 Curve Feature Extraction
3.1 Shape Context
3.2 Codebook Training
3.3 Quantization
4 Character Modeling by HMM
4.1 Preprocessing
4.2 DCE Algorithm
4.3 HMM Modeling
5 Experiments
5.1 Datasets and Experimental Setup
5.2 Optimal Parameters Setting
5.3 Results and Analysis
6 Conclusion
References
MusicMixer: Automatic DJ System Considering Beat and Latent Topic Similarity
1 Introduction
2 Related Work
2.1 Music Mixing and Playlist Generation
2.2 Topic Modeling
3 System Overview
4 Beat Similarity
5 Latent Topic Similarity
5.1 Topic Modeling
5.2 Calculation of Latent Topic Similarity
5.3 Evaluation
6 Mixing Songs
7 Discussion
7.1 Limitations
7.2 Applications of MusicMixer
7.3 Conclusion
References
Adaptive Synopsis of Non-Human Primates' Surveillance Video Based on Behavior Classification
1 Introduction
2 Related Work
3 Adaptive Synopsis
3.1 Preprocessing
3.2 Feature Extraction
3.3 Behavior Classification
3.4 Adaptive Synopsis
4 Experimental Results
4.1 Datasets
4.2 Behavior Classification
4.3 Video Synopsis
5 Conclusions and Future Work
References
A Packet Scheduling Method for Multimedia QoS Provisioning
1 Introduction
2 Related Works
3 Early Flow Discard Scheduling
4 EFD for a Full-Duplex Bottleneck Link
4.1 Simulation Methodology
4.2 EFD Internal Dynamics
4.3 Performation Evaluation
5 Conclusions
References
Robust Object Tracking Using Valid Fragments Selection
Abstract
1 Introduction
2 The Proposed Algorithm
2.1 DUFrags Selection Based on Discrimination and Uniqueness
2.2 V_DUFrags Selection Based on Harris-SIFT Filter and Spatial Constraint
2.3 Object Location Based on V_DUFrag Fusion
2.4 Feature Fusion Update
3 Experimental Results
3.1 DUFrags and V_DUFrags Selection
3.2 Tracking Results
3.3 Computational Cost
3.4 Limitations and Future Work
4 Conclusion
References
Special Session Poster Papers
Exploring Discriminative Views for 3D Object Retrieval
1 Introduction
2 Related Works
3 Proposed Method
3.1 Training
3.2 Querying
4 Experiments
5 Conclusion
References
What Catches Your Eyes as You Move Around? On the Discovery of Interesting Regions in the Street
1 Introduction
2 Interesting Regions
2.1 Attractive
2.2 Unique
2.3 Familiar
2.4 Spatio-Temporal Fusion
3 User Study
3.1 Data Set
3.2 Participants
3.3 Evaluation of Detected Interesting Regions
3.4 iNavi for Driving
4 Conclusions and Future Work
References
Bag Detection and Retrieval in Street Shots
1 Introduction
2 Related Work
3 Bag Detection with PC-CNN
3.1 Generating Bag Proposals with Selective Search
3.2 PC-CNN
3.3 Post Processing
4 Attribute Learning and Retrieval
5 Experiment
5.1 Dataset
5.2 Bag Detection
5.3 Attribute Learning and Bag Retrieval
6 Conclusions
References
TV Commercial Detection Using Success Based Locally Weighted Kernel Combination
1 Introduction
2 Audio-Visual Features
3 Success Based Locally Weighted Kernel Combination
4 Experimentation
4.1 TV News Commercial Dataset
4.2 Benchmark Datasets
4.3 Discussions
5 Conclusion
References
Frame-Wise Continuity-Based Video Summarization and Stretching
1 Introduction
2 Related Work
3 Frame-Wise Video Summarization
3.1 Frame-Wise Thinning Based on Visual Transition
3.2 Frame-Wise Thinning Out Based on Audio Transition
3.3 Frame-Wise Thinning Based on Audio-Visual Transition
4 Video Stretching via Frame Insertion
5 Subjective Experiment
6 Applications of the Proposed Method
7 Conclusion
References
Respiration Motion State Estimation on 4D CT Rib Cage Images
1 Introduction
2 Our Approach
2.1 Bone Segmentation
2.2 Motion State Estimation
3 Experiments
4 Conclusions
References
Location-Aware Image Classification
1 Introduction
2 Related Work
3 Approach
3.1 Local Feature Context
3.2 Latent SVM
4 Experiments
4.1 Experimental Setup
4.2 Experimental Results
5 Conclusions
References
Enhancement for Dust-Sand Storm Images
Abstract
1 Introduction
2 Proposed Method
2.1 Color Space Transformation
2.2 Color Cast Correction
2.3 Saturation Stretching
2.4 Detail Enhancement
3 Experimental Result and Comparison
4 Conclusion
References
Using Instagram Picture Features to Predict Users' Personality
1 Introduction
2 Related Work
3 Materials
4 Features
5 Results
5.1 Correlations
5.2 Personality Regressor
6 Discussion
7 Future Work and Limitations
8 Conclusion
References
Extracting Visual Knowledge from the Internet: Making Sense of Image Data
1 Introduction
2 Related Work
3 System Framework and Methods
3.1 Discovering Word Variations for the Given Concept
3.2 Purifying Noisy Word Variations
3.3 Purifying Noisy Images
3.4 Model Learning
4 Experiments and Analysis
5 Conclusion and Future Work
References
Ordering of Visual Descriptors in a Classifier Cascade Towards Improved Video Concept Detection
1 Introduction
2 Related Work
3 Cascade Construction with Pre-trained Classifiers
3.1 Cascade Architecture Overview
3.2 Problem Definition and Search Space
3.3 Problem Solution
4 Experiments
4.1 Dataset and Experimental Setup
4.2 Experimental Results
4.3 Computational Complexity
5 Conclusions
References
Spatial Constrained Fine-Grained Color Name for Person Re-identification
1 Introduction
2 Approach
2.1 Fine-Grained Color Name
2.2 Spatial Constrained Fine-Grained Color Name
2.3 Combination with Visual Features
3 Experiment
3.1 Dataset and Evaluation Protocol
3.2 Experimental Settings
3.3 Evaluation
3.4 Person Matching with Metric Learning
4 Conclusion
References
Dealing with Ambiguous Queries in Multimodal Video Retrieval
1 Introduction
2 Dealing with Ambiguous Queries
2.1 Considering Multiple Intents
2.2 Combining Intents
3 Cineast
4 Evaluation
4.1 Evaluation Procedure
4.2 Measurements
4.3 Results
5 Related Work
6 Conclusion
References
Collaborative Q-Learning Based Routing Control in Unstructured P2P Networks
1 Introduction
2 Proposed Method
3 Simulation Results
3.1 Performance Evaluation in Primitive Networks
3.2 Network Performance Evaluation Under High Churns
3.3 Network Performance Evaluation Under Heavy Workloads
4 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken