Image Analysis and Processing — ICIAP 2015 [[electronic resource] ] : 18th International Conference, Genoa, Italy, September 7-11, 2015, Proceedings, Part I / / edited by Vittorio Murino, Enrico Puppo

• Pubblicazione: Cham : , : Springer International Publishing : , : Imprint : Springer, , 2015
• Edizione: 1st ed. 2015.
• Descrizione fisica: 1 online resource (XXVI, 721 p. 279 illus.)
• Disciplina: 621.367
• Soggetto topico: Optical data processing; Pattern recognition; Artificial intelligence; Algorithms; Computer graphics; Image Processing and Computer Vision; Pattern Recognition; Artificial Intelligence; Algorithm Analysis and Problem Complexity; Computer Graphics
• Persona (resp. second.): MurinoVittorio; PuppoEnrico
• Note generali: Bibliographic Level Mode of Issuance: Monograph
• Nota di bibliografia: Includes bibliographical references and index.
• Nota di contenuto: Intro -- Preface -- Organization -- Contents - Part I -- Contents - Part II -- Pattern Recognition and Machine Learning -- Transfer Learning Through Greedy Subset Selection -- 1 Introduction -- 2 Related Work -- 3 Transfer Learning Through Subset Selection -- 4 Greedy Algorithm for k-Source Selection -- 5 Experiments -- 6 Conclusions -- References -- MEG: Multi-Expert Gender Classification from Face Images in a Demographics-Balanced Dataset -- 1 Introduction -- 2 The Pool of Experts -- 2.1 Score Generation by Likelihood Ratio (LR) -- 2.2 Score Generation by Support Vector Machines (SVM) -- 3 Fusion Strategies -- 4 The Image Dataset -- 5 Experiments and Results -- 6 Conclusions -- References -- An Edge-Based Matching Kernel Through Discrete-Time Quantum Walks -- 1 Introduction -- 2 Preliminary Concepts -- 2.1 Directed Line Graphs -- 2.2 Discrete-Time Quantum Walks -- 2.3 The Relation to Perron-Frobenius Operators -- 2.4 Depth-Based Representations Based on Quantum Walks -- 3 An Edge-Based Matching Kernel for Graphs -- 3.1 Edge-Based Matching Through Discrete-Time Quantum Walks -- 3.2 An Edge-Based Matching Kernel -- 4 Experimental Results -- 5 Conclusion -- References -- Implicit Boundary Learning for Connectomics -- 1 Introduction -- 2 Method -- 2.1 Finding Labels to Optimize Segmentation -- 2.2 Difference to Learn Membrane -- 3 Experiments and Results -- 4 Conclusion -- References -- A Parzen-Based Distance Between Probability Measures as an Alternative of Summary Statistics in Approximate Bayesian Computation -- 1 Introduction -- 2 Approximate Bayesian Computation Based on Kernel Embeddings -- 2.1 Short Summary on ABC Methods -- 2.2 Metrics between Probability Measures by Using Kernels -- 2.3 Kernel Embeddings as Summary Statistics for ABC -- 2.4 Extension to ABC SMC -- 3 Experimental Setup -- 3.1 Datasets -- 3.2 Validation -- 3.3 Procedure. ; 4 Results and Discussion -- 4.1 Results from Synthetic Data -- 4.2 Results from Nonlinear Ecological Dynamic System -- 5 Conclusions -- References -- Unsupervised Classification of Raw Full-Waveform Airborne Lidar Data by Self Organizing Maps -- 1 Introduction -- 2 Methodology -- 3 Experiments and Results -- 4 Discussion and Conclusion -- References -- Fitting Multiple Models via Density Analysis in Tanimoto Space -- 1 Introduction -- 2 Method -- 2.1 Preference Trick -- 2.2 Density Based Analysis for Model Extraction -- 3 Experimental Evaluation -- 4 Conclusion -- References -- Bag of Graphs with Geometric Relationships Among Trajectories for Better Human Action Recognition -- 1 Introduction -- 2 Related Work -- 3 Trajectory Extraction -- 4 Description of the Proposed Approach -- 4.1 Extraction of the Graph Descriptors -- 4.2 Bag of Graphs -- 5 Experimental Study -- 5.1 Action Recognition Datasets -- 5.2 Results -- 6 Conclusion -- References -- Have a SNAK. Encoding Spatial Information with the Spatial Non-alignment Kernel -- 1 Introduction -- 2 Related Work -- 3 Spatial Non-Alignment Kernel -- 3.1 Translation and Size Invariance -- 4 Experiments -- 4.1 Data Sets Description -- 4.2 Implementation and Evaluation Procedure -- 4.3 Parameter Tuning -- 4.4 Pascal VOC Experiment -- 4.5 Birds Experiment -- 5 Conclusion and Future Work -- References -- Convolved Multi-output Gaussian Processes for Semi-Supervised Learning -- 1 Introduction -- 2 Multi-output Gaussian Processes -- 3 Semi-Supervised Learning for Multi-output GPs -- 3.1 Graph Regularization -- 3.2 The EM Algorithm -- 4 Experimental Setup -- 4.1 Databases -- 4.2 Procedure -- 5 Results and Discussion -- 6 Conclusions and Future Work -- References -- Volcano-Seismic Events Classification Using Document Classification Strategies -- 1 Introduction -- 2 The Proposed Approach. ; 2.1 Dictionary Building -- 2.2 Event Description -- 2.3 From Documents to Feature Vectors -- 2.4 Document Classification -- 3 Experiments -- 3.1 Experimental Details -- 3.2 Results and Comparison with other Methods -- 3.3 Effect of the Parameters -- 3.4 Interpretability -- 4 Conclusions -- References -- Unsupervised Feature Selection by Graph Optimization -- 1 Introduction -- 2 A Brief Review of Graph-Based Unsupervised Feature Selection Methods -- 3 Unsupervised Feature Selection by Graph Optimization -- 4 Optimization Algorithm for Problem (5) -- 5 Experiments and Comparisons -- 6 Conclusion -- References -- Gait Recognition Robust to Speed Transition Using Mutual Subspace Method -- 1 Introduction -- 2 Gait Recognition Using Mutual Subspace Method -- 2.1 Mutual Subspace Method -- 2.2 Gait Recognition Using an Image Set-Based Matching -- 3 Experiments -- 3.1 Experimental Settings -- Datasets. -- Evaluation Setting. -- 3.2 Evaluation -- 4 Conclusions -- References -- Path-Based Dominant-Set Clustering -- 1 Introduction -- 2 Path-Based Dominant Sets -- 2.1 Dominant Set Clustering -- 2.2 Using Path-Based Similarity -- 3 Experiments -- 3.1 Synthetic Data Clustering -- 3.2 Experiments on Real-World Data -- 4 Conclusion -- References -- Global and Local Gaussian Process for Multioutput and Treed Data -- 1 Introduction -- 2 Materials and Methods -- 2.1 Gaussian Process -- 2.2 Multiple Output Gaussian Process -- 2.3 Global and Local Multi Output Treed GP -- 3 Results -- 3.1 Validation and Error Measures -- 3.2 Results over Simulated Data -- 3.3 Results over Real Data -- 4 Conclusions -- References -- BRISK Local Descriptors for Heavily Occluded Ball Recognition -- 1 Introduction -- 2 Ball Recognition Evaluation -- 2.1 Moving Object Segmentation -- 2.2 Circle Detection -- 2.3 Feature Extraction -- 2.4 Supervised Classifier -- 2.5 Experimental Setup. ; 3 Experimental Results -- 4 Conclusion and Future Work -- References -- Neighborhood Selection for Dimensionality Reduction -- 1 Introduction -- 2 Algorithm -- 3 Experimental Results and Future Works -- References -- Crowdsearching Training Sets for Image Classification -- 1 Introduction -- 2 State of the Art -- 3 Proposed Methodology -- 4 Experiments and Results -- 4.1 Comparison Against the Semantic Trainer -- 4.2 Generalization Capability -- 4.3 Comparison against OPTIMOL -- 5 Conclusion -- References -- The Color of Smiling: Computational Synaesthesia of Facial Expressions -- 1 Introduction -- 2 Background and Rationales -- 3 Methods -- 4 From Face Expression to Mood -- 5 The Color of Mood -- 6 Conclusion and Further Outlooks -- References -- Learning Texture Image Prior for Super Resolution Using Restricted Boltzmann Machine -- 1 Introduction -- 2 Learning Texture Image Prior -- 2.1 Modeling Individual Texture with the RBM -- 2.2 Implicit Mixture Modeling for Multiple Textures -- 3 A MAP Framework for Super Resolution -- 3.1 Image Formation Modeling -- 3.2 Embedding Multiple Texture Prior in the MAP Framework -- 4 Experiment -- 4.1 Dataset and Learning cRBM -- 4.2 Multiple Texture Super Resolution -- 5 Conclusion -- References -- GRUNTS: Graph Representation for UNsupervised Temporal Segmentation -- 1 Introduction -- 2 Graph Representation -- 2.1 Skeleton -- 2.2 Polygonal Approximation -- 2.3 Building the Graph -- 3 The Frame Kernel Matrix -- 4 Aligned Cluster Analysis (ACA) -- 5 The GRUNTS Algorithm -- 6 Experiments -- 6.1 Weizmann Dataset -- 6.2 KTH Dataset -- 6.3 MSR Action3D Dataset -- 7 Conclusions -- References -- A Strict Pyramidal Deep Neural Network for Action Recognition -- 1 Introduction -- 2 3DPyraNet -- 2.1 PyraNet -- 2.2 Weighting Scheme -- 2.3 Proposed Architecture -- 3 Results and Discussion -- 4 Conclusions -- References. ; Nerve Localization by Machine Learning Framework with New Feature Selection Algorithm -- 1 Introduction -- 2 Nerve Detection System -- 2.1 Pre-processing -- 2.2 Feature Extraction -- 2.3 Feature Selection -- 2.4 Proposed Feature Selection Algorithm -- 2.5 Classification -- 3 Experiment Results -- 4 Conclusion -- References -- Human Tracking Using a Top-Down and Knowledge Based Approach -- 1 Introduction -- 2 Top-Down Approach -- 3 Knowledge Related to Tracking -- 3.1 Scene Knowledge -- 3.2 Human Trajectories Knowledge -- 3.3 From Knowledge to Model -- 4 Preliminary Results and Discussions -- 5 Conclusion and Future Directions -- References -- Shape Analysis and 3D Computer Vision -- Fuzzy ``Along'' Spatial Relation in 3D. Application to Anatomical Structures in Maxillofacial CBCT -- 1 Introduction -- 2 General Approach to Define ``Along'' -- 3 Inter-Objects Region -- 4 Elongation Measure -- 4.1 Common Definition -- 4.2 Alongness Degree Based on Object Boundaries -- 4.3 Including Distance Information -- 4.4 Using Distance to Take into Account Only Close Parts -- 5 Extension to Fuzzy Objects -- 6 Experimental Results -- 7 Conclusion -- References -- Compression and Querying of Arbitrary Geodesic Distances -- 1 Introduction -- 2 Related Work -- 3 Algorithm -- 3.1 Patch Subdivision -- 3.2 Geodesic Precomputation -- 3.3 Graph Pruning -- 3.4 Query Step -- 4 Results -- 4.1 Parameters Tuning -- 4.2 Preprocessing Time -- 4.3 Speedup and Comparisons -- 5 Conclusions -- References -- Comparing Persistence Diagrams Through Complex Vectors -- 1 Introduction -- 2 Preliminaries -- 3 Persistence Diagrams vs Complex Vectors -- 4 Experimental Results -- References -- Pop-up Modelling of Hazy Scenes -- 1 Introduction -- 2 Related Work -- 2.1 Interactive 3D Modelling -- 2.2 Automatic 3D Modelling -- 3 Model Construction -- 3.1 Dehazing. ; 3.2 Clustering and Plane-fitting.
• Sommario/riassunto: The two-volume set LNCS 9279 and 9280 constitutes the refereed proceedings of the 18th International Conference on Image Analysis and Processing, ICIAP 2015, held in Genoa, Italy, in September 2015. The 129 papers presented were carefully reviewed and selected from 231 submissions. The papers are organized in the following seven topical sections: video analysis and understanding, multiview geometry and 3D computer vision, pattern recognition and machine learning, image analysis, detection and recognition, shape analysis and modeling, multimedia, and biomedical applications.
• Titolo autorizzato: Image Analysis and Processing — ICIAP 2015
• ISBN: 3-319-23231-2
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 996200363303316
• Lo trovi qui: Univ. di Salerno
• Serie: Image Processing, Computer Vision, Pattern Recognition, and Graphics ; ; 9279