Note: Unterschiede zwischen dem gedruckten Dokument und der elektronischen Ressource können nicht ausgeschlossen werden , Förderkennzeichen BMBF 13EX1015A. - Verbund-Nr. 01079002 , Systemvoraussetzungen: Acrobat reader.

Note: Förderkennzeichen BMBF 13EX1212G. - Verbund-Nummer 01079050 , Paralleltitel dem englischen Berichtsblatt entnommen , Autoren dem Berichtsblatt entnommen , Unterschiede zwischen dem gedruckten Dokument und der elektronischen Ressource können nicht ausgeschlossen werden , Zusammenfassungen in deutscher und englischer Sprache

Note: Förderkennzeichen BMBF 01UG1236A. - Verbund-Nummer 01125345 , Unterschiede zwischen dem gedruckten Dokument und der elektronischen Ressource können nicht ausgeschlossen werden

Note: Förderkennzeichen BMBF 13GW0031A. - Verbund-Nummer 01145746 , Paralleltitel dem englischen Berichtsblatt entnommen , Autoren dem Berichtsblatt entnommen , Unterschiede zwischen dem gedruckten Dokument und der elektronischen Ressource können nicht ausgeschlossen werden , Zusammenfassungen in deutscher und englischer Sprache

Note: Description based on publisher supplied metadata and other sources

Note: Intro -- Preface -- Organization -- Awards 2013 -- Contents -- Variational Models for Depth and Flow -- Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 4 Light Field Scene Flow Model -- 4.1 Data Fidelity Term -- 4.2 Regularization Term -- 4.3 Discretization -- 4.4 Optimization -- 5 Experimental Results -- 5.1 Synthetic Experiments -- 5.2 Real World Experiments -- 6 Conclusion -- References -- Introducing More Physics into Variational Depth--from--Defocus -- 1 Introduction -- 2 Image Formation Models -- 2.1 Image Formation with a Thin Lens -- 2.2 Spatially Variant Convolution -- 2.3 Approximation of the PSF -- 2.4 Our Modification -- 3 Variational Formulation -- 3.1 Variational Model -- 3.2 Minimisation -- 3.3 Discretisation and Implementation -- 4 Experiments -- 5 Conclusions -- References -- Reconstruction -- High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth -- 1 Introduction -- 2 Related Work -- 3 Processing Pipeline -- 4 Stereo Experiments -- 5 Conclusion -- References -- Semi-Global Matching: A Principled Derivation in Terms of Message Passing -- 1 Introduction -- 2 Notation and Background -- 2.1 Markov Random Fields -- 2.2 Min-Sum Belief Propagation (BP) -- 2.3 Tree-Reweighted Message-Passing (TRW-T) -- 2.4 Semi-Global Matching (SGM) -- 3 Synthesis -- 3.1 Relation between SGM and BP on Union Jack -- 3.2 Relation between SGM and TRW-T -- 4 Implications -- 5 Conclusion -- References -- Submap-Based Bundle Adjustment for 3D Reconstruction from RGB-D Data -- 1 Introduction -- 2 Related Work -- 3 3D Object Reconstruction Pipeline -- 4 Full Bundle Adjustment for RGB-D Sensors -- 5 Efficient Bundle Adjustment for RGB-D Sensors Using Submapping -- 6 Evaluation and Experimental Results -- 7 Conclusion and Future Work -- References. , Bio-informatics -- A Hierarchical Bayesian Approach for Unsupervised Cell Phenotype Clustering -- 1 Introduction -- 2 Previous Works -- 3 Wordless HDP-HMM -- 3.1 Modeling Principles -- 3.2 Inference -- 4 Experiments and Results -- 4.1 Data-Sets -- 4.2 Experimental Set-Up and Evaluation -- 4.3 Results and Discussion -- 5 Conclusions -- References -- Information Bottleneck for Pathway-Centric Gene Expression Analysis -- 1 Introduction -- 2 Information Bottleneck (IB) -- 2.1 Gaussian Information Bottleneck (GIB) -- 2.2 Meta-Gaussian Information Bottleneck (MGIB) -- 3 Extensions for Pathway Analysis -- 3.1 Mixed Data and Missing Values -- 3.2 Relation to PCA -- 3.3 Irrelevance Variables -- 4 Experiments -- 4.1 Comparison to Classic Pathifier Using PCA -- 4.2 Analysis with Relevance and Irrelevance Variables -- 5 Conclusion -- References -- Deep Learning and Segmentation -- Convolutional Decision Trees for Feature Learning and Segmentation -- 1 Introduction -- 2 Related Work -- 2.1 Feature Learning -- 2.2 Binary Decision Tree -- 3 Proposed Method -- 3.1 Information Gain -- 3.2 Regularization -- 3.3 Optimization -- 4 Decision Trees -- 5 Experiments -- 5.1 Experimental Settings -- 5.2 Drosophila VNC Dataset -- 5.3 Weizmann Horse Dataset -- 6 Conclusion -- References -- A Deep Variational Model for Image Segmentation -- 1 Introduction -- 2 CNNs and Variational Image Segmentation -- 3 Joint Training -- 4 Experiments -- 5 Conclusion -- References -- Feature Computation -- Robust PCA: Optimization of the Robust Reconstruction Error Over the Stiefel Manifold -- 1 Introduction -- 2 Robust PCA -- 3 TRPCA: Minimizing Trimmed Reconstruction Error on the Stiefel Manifold -- 3.1 Reformulation and First Properties -- 3.2 Minimization Algorithm -- 3.3 Complexity and Discussion -- 4 Experiments -- 4.1 Synthetic Data Sets -- 4.2 Partially Synthetic Data Set. , 4.3 Background Modeling and Subtraction -- 5 Conclusion -- References -- An O(n logn) Cutting Plane Algorithm for Structured Output Ranking -- 1 Introduction -- 2 Structured Output Ranking -- 3 O(n logn) Cutting Plane Algorithm -- 3.1 Complexity -- 4 Generalization Bounds -- 5 Experimental Results -- 6 Discussion -- 7 Conclusions -- References -- Exemplar-Specific Patch Features for Fine-Grained Recognition -- 1 Introduction -- 2 Related Work -- 3 Discovering Mid-Level Patch Representations -- 4 Exemplar-Specific Mid-level Features -- 5 Experiments -- 5.1 Evaluation of Global Patch Discovery -- 5.2 Evaluation of Exemplar-Specific Representations -- 6 Conclusions -- References -- Video Interpretation -- Motion Segmentation with Weak Labeling Priors -- 1 Introduction -- 2 Integrating Prior Knowledge into Motion Segmentation -- 3 Tracking Based on Segmentation -- 4 Experimental Results and Discussion -- 4.1 Performance on Videos of Infants -- 4.2 Comparison with Standard Benchmarks -- 4.3 Performance on Cerebral Palsy Detection -- 5 Conclusions -- References -- Object-Level Priors for Stixel Generation -- 1 Introduction -- 1.1 Related Work -- 2 The Stixel World -- 2.1 Probabilistic Reformulation -- 3 Incorporating Priors -- 3.1 Generic Prior Model -- 3.2 Probabilistic Formulation -- 4 Experimental Results -- 4.1 Priors -- 4.2 Dataset -- 4.3 Segmentation Accuracy -- 4.4 Precision -- 4.5 Freespace Estimation -- 5 Conclusion -- References -- Coherent Multi-sentence Video Description with Variable Level of Detail -- 1 Introduction -- 2 Generating Consistent Multi-sentence Video Descriptions at Multiple Levels of Detail -- 2.1 Analysis of Human Video Descriptions at Multiple Levels of Detail -- 2.2 Multi-sentence Video Descriptions -- 2.3 Multi-Level Video Descriptions -- 3 Improving Visual Features -- 4 Generating Natural Descriptions -- 5 Evaluation. , 5.1 Visual Recognition -- 5.2 Multi-sentence Generation -- 5.3 Multi-Level Generation -- 6 Conclusion -- References -- Segmentation and Labeling -- Asymmetric Cuts: Joint Image Labeling and Partitioning -- 1 Introduction -- 2 Related Work -- 3 Multi-way Cut Formulation -- 4 The Asymmetric Multi-way Cut Formulation -- 4.1 Formulation Within the Binary Edge Labeling Framework -- 4.2 Formulation as a Node Labeling Problem -- 4.3 Labeling of Regions and Boundaries -- 5 Experiments -- 6 Conclusion -- References -- Mind the Gap: Modeling Local and Global Context in (Road) Networks -- 1 Introduction -- 2 Related Work -- 3 Model -- 3.1 Context-Aware Road Scores -- 3.2 Road Likelihoods -- 3.3 Sampling of Road Candidates -- 3.4 Candidate Selection -- 4 Experimental Results -- 4.1 Evaluation Metrics -- 4.2 Results -- 5 Conclusions and Future Work -- References -- Image Processing and Analysis -- Guided Image Super-Resolution: A New Technique for Photogeometric Super-Resolution in Hybrid 3-D Range Imaging -- 1 Introduction -- 1.1 Related Work -- 1.2 Contribution -- 2 Photogeometric Super-Resolution Model -- 2.1 Data Fidelity Term -- 2.2 Smoothness Regularization -- 2.3 Interdependence Regularization -- 3 Numerical Optimization Algorithm -- 4 Experiments and Results -- 4.1 Synthetic Hybrid Range Data -- 4.2 Microsoft's Kinect Datasets -- 5 Discussion -- 6 Conclusion and Future Work -- References -- Image Descriptors Based on Curvature Histograms -- 1 Introduction -- 2 Related Work -- 3 Curvature -- 3.1 Dense Vector Curvature on Images -- 3.2 Feature Binning -- 4 Experiments -- 4.1 Image Classification -- 4.2 Object Detection -- 4.3 Descriptor Matching -- 5 Conclusions -- References -- Human Pose and People Tracking -- Test-Time Adaptation for 3D Human Pose Estimation -- 1 Introduction and Related Work -- 2 Multi-view Pictorial Structures -- 3 Test-Time Adaptation. , 3.1 Confident Examples Mining -- 3.2 2D Model Refinement -- 4 Experiments -- 5 Conclusion -- References -- Efficient Multiple People Tracking Using Minimum Cost Arborescences -- 1 Introduction -- 1.1 Related Work -- 1.2 Contributions -- 2 Tracklet Creation -- 2.1 Tracklet Creation with Arborescences -- 2.2 Using Bifurcations to Detect Ambiguities -- 3 Implementation Details -- 4 Experiments -- 4.1 Influence of the Parameters -- 4.2 Runtime -- 4.3 Tracking Performance -- 5 Conclusion -- References -- Capturing Hand Motion with an RGB-D Sensor, Fusing a Generative Model with Salient Points -- 1 Introduction -- 2 Related Work -- 3 Tracking Method -- 3.1 Hand Model -- 3.2 Optimization -- 4 Experimental Evaluation -- 4.1 Pipeline Components -- 4.2 Comparison to State-of-the-Art -- 5 Conclusion -- References -- Interpolation and Inpainting -- Flow and Color Inpainting for Video Completion -- 1 Introduction -- 1.1 Related Work -- 1.2 Contributions -- 2 Interactive Mask-Definition -- 2.1 Scribble Relocation via Optical Flow -- 2.2 Segmentation According to Nieuwenhuis and Cremers -- 3 Flow and Color Inpainting for Video Completion -- 3.1 Inpainted Flow for Temporal Coherence -- 3.2 Flow Inpainting -- 3.3 Exemplar-Based Inpainting -- 3.4 Flow Preserving Distance Function -- 3.5 Overview of the Algorithm -- 4 Experiments and Results -- 4.1 Implementation and Runtime -- 5 Conclusion -- References -- Spatial and Temporal Interpolation of Multi-view Image Sequences -- 1 Introduction -- 1.1 Related Work -- 1.2 Contribution -- 2 Image Interpolation -- 2.1 Spatial Interpolation -- 2.2 Temporal Interpolation -- 2.3 Image Warping and Blending -- 3 Results -- 4 Conclusion -- References -- Pose Normalization for Eye Gaze Estimation and Facial Attribute Description from Still Images -- 1 Introduction -- 1.1 Prior Work -- 2 Methods -- 2.1 Face Model. , 2.2 Pose Normalization.

Note: Description based on publisher supplied metadata and other sources