Technical Program
Monday, September 3 9:30 - 11:00
T01-A: Signal Processing and Optimization for Ultra Reliable Low Latency Communications
Eduard Axel Jorswieck (TU Dresden, Germany); Muhammad Ali Imran (University of Glasgow, UK); Muhammad Majid Butt (University of Glasgow, UK)
Room QT02-A: Light field image processing
Neus Sabater (Technicolor, France); Christine Guillemot (Institut National de Recherche en Informatique et Automatique, France)
Room GT03-A: Wild Patterns: Ten Years after the rise of Adversarial Machine Learning
Battista Biggio (University of Cagliari, Italy); Fabio Roli (University of Cagliari, Italy)
Room AT04-A: Deep Learning for Image and Video Processing
A. Murat Tekalp (Koc University, Turkey)
Room PininfarinaT05-A: Signal Processing for Autonomous and Self-Aware systems
Carlo Regazzoni (University of Genova, Italy); Lucio Marcenaro (University of Genova, Italy)
Room HMonday, September 3 11:00 - 11:30
Communication Break
Monday, September 3 11:30 - 13:00
T01-B : Signal Processing and Optimization for Ultra Reliable Low Latency Communications
Eduard Axel Jorswieck (TU Dresden, Germany); Muhammad Ali Imran (University of Glasgow, UK); Muhammad Majid Butt (University of Glasgow, UK)
Room QT02-B: Light field image processing
Neus Sabater (Technicolor, France); Christine Guillemot (Institut National de Recherche en Informatique et Automatique, France)
Room GT03-B: A Wild Patterns: Ten Years after the rise of Adversarial Machine Learning
Battista Biggio (University of Cagliari, Italy); Fabio Roli (University of Cagliari, Italy)
Room AT04-B: Deep Learning for Image and Video Processing
A. Murat Tekalp (Koc University, Turkey)
Room PininfarinaT05-B: Signal Processing for Autonomous and Self-Aware systems
Carlo Regazzoni (University of Genova, Italy); Lucio Marcenaro (University of Genova, Italy)
Room HMonday, September 3 13:00 - 14:30
Lunch
Monday, September 3 14:30 - 16:00
T06-A: Vertical-Oriented End-to-End Orchestration in 5G Networks: Modeling, Optimization, Implementation, and Verification
Vincenzo Sciancalepore (NEC Laboratories Europe, Germany); Merouane Debbah (CentraleSupélec, France) Alessio Zappone (University of Cassino and Southern Lazio, Italy); Marco Di Renzo (CentraleSupélec, France)
Room GT07-A: Signal Processing Methods for Light Field 3D Displays
Robert Bregovic (Tampere University of Technology, Finland); Erdem Sahin (Tampere University of Technology, Finland); Atanas Gotchev (Tampere University of Technology, Finland)
Room AT08-A: Random Matrix Advances in Machine Learning and Neural Nets
Romain Couillet (CentraleSupélec, France); Zhenyu Liao (CentraleSupélec, France); Xiaoyi Mai (CentraleSupélec, France)
Room QT09-A: Robust Covariance and Subspace Learning: dealing with high-dimensionality and small sample support
Arnaud Breloy (University Paris Nanterre, France); Guillaume Ginolhac (Université Savoie Mont Blanc, France); Frédéric Pascal (University Paris-Saclay, France)
Room HT10-A: Deep learning for Multimedia Forensics
Luisa Verdoliva (University Federico II of Naples, Italy)
Room PininfarinaMonday, September 3 16:00 - 16:30
Communication Break
Monday, September 3 16:30 - 18:00
T06-B: Vertical-Oriented End-to-End Orchestration in 5G Networks: Modeling, Optimization, Implementation, and Verification
Vincenzo Sciancalepore (NEC Laboratories Europe, Germany); Merouane Debbah (CentraleSupélec, France) Alessio Zappone (University of Cassino and Southern Lazio, Italy); Marco Di Renzo (CentraleSupélec, France)
Room GT07-B: Signal Processing Methods for Light Field 3D Displays
Robert Bregovic (Tampere University of Technology, Finland); Erdem Sahin (Tampere University of Technology, Finland); Atanas Gotchev (Tampere University of Technology, Finland)
Room AT08-B: Random Matrix Advances in Machine Learning and Neural Nets
Romain Couillet (CentraleSupélec, France); Zhenyu Liao (CentraleSupélec, France); Xiaoyi Mai (CentraleSupélec, France)
Room QT09-B: A Robust Covariance and Subspace Learning: dealing with high-dimensionality and small sample support
Arnaud Breloy (University Paris Nanterre, France); Guillaume Ginolhac (Université Savoie Mont Blanc, France); Frédéric Pascal (University Paris-Saclay, France)
Room HT10-B: A Deep learning for Multimedia Forensics
Luisa Verdoliva (University Federico II of Naples, Italy)
Room PininfarinaMonday, September 3 19:00 - 22:00
Welcome Reception
Tuesday, September 4 8:30 - 9:30
Opening and Award Ceremony
Tuesday, September 4 9:30 - 10:30
PL1: Coalitional Multimedia Signal Processing in Smart Device Networks
Marc Moonen (KU Leuven, Belgium)
Room: AuditoriumTuesday, September 4 10:30 - 11:00
Communication Break
Tuesday, September 4 11:00 - 13:00
AASP-L1: Acoustic sensor array processing
Room Pininfarina
Chair: Sharon Gannot (Bar-Ilan University, Israel)
- 11:00 Distant Noise Reduction Based on Multi-delay Noise Model Using Distributed Microphone Array
- 11:20 Joint Beamforming and Echo Cancellation Combining QRD Based Multichannel AEC and MVDR for Reducing Noise and Non-Linear Echo
- 11:40 Steerable Circular Differential Microphone Arrays
- 12:00 Speech Enhancement by Classification of Noisy Signals Decomposed Using NMF and Wiener Filtering
- 12:20 Joint Identification and Localization of a Speaker in Adverse Conditions Using a Microphone Array
- 12:40 Acoustic Beamforming in Front of a Reflective Plane
BSIP-P1: Biomedical signal and image processing
Chair: Francoise Peyrin (Universite de Lyon INSA Lyon & Inserm U1206 UMR CNRS 5220, France)
- Velocity Variability in MRI Phase-Contrast
- Analysis Vs Synthesis-based Regularization for Combined Compressed Sensing and Parallel MRI Reconstruction at 7 Tesla
- A Hierarchical Ensemble Classifier for Multilabel Segmentation of Fat-Water MR Images
- High Frequency Noise Detection and Handling in ECG Signals
- Denoising Phonocardiogram Signals with Non-negative Matrix Factorization Informed by Synchronous Electrocardiogram
- Fusion of EEG and fMRI via Soft Coupled Tensor Decompositions
- DeepMQ: A Deep Learning Approach Based Myelin Quantification in Microscopic Fluorescence Images
- Ensemble Learning for Detection of Short Episodes of Atrial Fibrillation
- An ADMM Algorithm for Constrained Material Decomposition in Spectral CT
- A Resolution Enhancement Technique for Ultrafast Coded Medical Ultrasound
- An Enhanced Chirp Modulated Golay Code for Ultrasound Diverging Wave Compounding
- Beamforming and Blind Source Separation Have a Complementary Effect in Reducing Tonic Cranial Muscle Contamination of Scalp Measurements
- Multi-task Feature Learning for EEG-based Emotion Recognition Using Group Nonnegative Matrix Factorization
- A Dynamic Model of Synthetic Resting-State Brain Hemodynamics
- How Many Channels are Enough? Evaluation of Tonic Cranial Muscle Artefact Reduction Using ICA with Different Numbers of EEG Channels
- Classification of EEG Signals Based on Mean-Square Error Optimal Time-Frequency Features
GSP-L1: Graph Signal Processing
Room: Auditorium
Chair: Cédric Richard (Université de Nice Sophia-Antipolis, France)
- 11:00 A Preconditioned Graph Diffusion LMS for Adaptive Graph Signal Processing
- 11:20 Spectral MAB for Unknown Graph Processes
- 11:40 Blind Identification of Invertible Graph Filters with Multiple Sparse Inputs
- 12:00 A Novel Method for Sampling Bandlimited Graph Signals
- 12:20 Online Recovery of Time-varying Signals Defined over Dynamic Graphs
- 12:40 Extreme Learning Machine for Graph Signal Processing
IVM-P1: Image, video and multimedia processing
Chair: Riccardo Mazzon (Queen Mary University of London, United Kingdom (Great Britain))
- Shearlet-based Loop Filter
- A Forward-Looking Sonar Progressive Coding Using Morphological Skeleton Representation
- Adaptive Video Encoding for Time-Constrained Compression and Delivery
- A Lossless Image Coding Method Based on Probability Model Optimization
- Coding Gain Optimized 8-Point DST with Fast Algorithm for Intra-frames in Video Coding
- Noisy Encrypted Image Correction Based on Shannon Entropy Measurement in Pixel Blocks of Very Small Size
- Visual Saliency Guided High Dynamic Range Image Compression
- End-to-End Real-Time ROI-based Encryption in HEVC Videos
SPPS-P1: Signal processing for Power Systems
Chair: Levent Eren (Izmir University of Economics, Turkey)
- Iterative Weighted Least Squares Frequency Estimation for Harmonic Sinusoidal Signal in Power Systems
- Renewable Energy Optimization with Centralized and Distributed Generation
- Deep Residual Neural Network for EMI Event Classification Using Bispectrum Representations
- Cooperative Renewable Energy Management with Distributed Generation
- Motor Condition Monitoring by Empirical Wavelet Transform
SS-L1: Special Session on Full Parallax Imaging
Room Q
Chair: Atanas Gotchev (Tampere University of Technology, Finland)
Organizers: Atanas Gotchev and Mårten Sjöström
- 11:00 A Novel Kinect V2 Registration Method Using Color and Deep Geometry Descriptors
- 11:20 Area-Based Depth Estimation for Monochromatic Feature-Sparse Orthographic Capture
- 11:40 On the Possibility to Achieve 6-DoF for 360 Video Using Divergent Multi-View Content
- 12:00 Continuous Refocusing for Integral Microscopy with Fourier Plane Recording
- 12:20 Modeling the Pairwise Disparities in High Density Camera Arrays
- 12:40 Multi-Shot Single Sensor Light Field Camera Using a Color Coded Mask
SS-L2: Special Session on Recent Advances in Immersive Imaging Technologies
Room G
Chair: Martin Alain (Trinitry College Dublin, Ireland)
Organizers: Martin Alain, Cagri Ozcinar, and Sebastian Knorr
- 11:00 Low Cost Setup for High Resolution Multiview Panorama Recording and Registration
- 11:20 Lenslet Light Field Panorama Creation: a Sub-Aperture Image Stitching Approach
- 11:40 Light-fields of Circular Camera Arrays
- 12:00 A Rendering Solution to Display Light Field in Virtual Reality
- 12:20 Multi-stereo Matching for Light Field Camera Arrays
- 12:40 Steered Mixture-of-Experts Approximation of Spherical Image Data
SSP-L1: Statistical Signal Processing
Room A
Chair: Fulvio Gini (University of Pisa, Italy)
- 11:00 A Fresh Look at the Semiparametric Cramér-Rao Bound
- 11:20 Bayesian Multi-Class Covariance Matrix Filtering for Adaptive Environment Learning
- 11:40 Generalized Conditional Maximum Likelihood Estimators in the Large Sample Regime
- 12:00 A Novel Recursive Bayesian Weighted Instrumental Variable Estimator for 3D Bearings-only TMA
- 12:20 Performance Bounds for Change Point and Time Delay Estimation
- 12:40 Consistent Spectral Methods for Dimensionality Reduction
Tuesday, September 4 13:00 - 14:30
Lunch
Tuesday, September 4 13:15 - 14:15
MathWorks Workshop
A practical approach to Deep Learning with MATLAB
Stefano Olivieri (Academia Group - MathWorks)
Room ASpringer Workshop
Alle-Jan van der Veen (TU Delft, Director of Publications of Eurasip), Angelina Wagner (Springer)
Room GTuesday, September 4 14:30 - 16:10
BSIP-L1: Biomedical signal and image processing
Room: Auditorium
Chair: Maria Joao Rendas (Centre Nacional Recherche Scientifique & Laboratoire I3S, France)
- 14:30 Refined Generalized Multivariate Multiscale Fuzzy Entropy: a Preliminary Study on Multichannel Physiological Complexity During Postural Changes
- 14:50 Building a Tensor Framework for the Analysis and Classification of Steady-State Visual Evoked Potentials in Children
- 15:10 Comparison of Parametric and Non-Parametric Population Modelling of Sport Performances
- 15:30 Constrained Particle Filter for Improving Kinect Based Measurements
- 15:50 Subspace Classification of Human Gait Using Radar Micro-Doppler Signatures
NLSP-L1: Nonlinear Signal Processing
Room Pininfarina
Chair: Alberto Carini (University of Urbino, Italy)
- 14:30 A New Stream Cipher Based on Nonlinear Dynamic System
- 14:50 Species Related Gas Tracking in Distribution Grids
- 15:10 Fast Algorithms for the Schur-Type Nonlinear Parametrization of Higher-Order Stochastic Processes
- 15:30 Efficient Generation of Random Signals with Prescribed Probability Distribution and Spectral Bandwidth via Ergodic Transformations
- 15:50 Combined Sparse Regularization for Nonlinear Adaptive Filters
SENS-L1: Sensor Arrays in Source Localization
Room G
Chair: Walter Kellermann (University Erlangen-Nuremberg, Germany)
- 14:30 Localization of Near-Field Signals Based on Linear Prediction and Oblique Projection Operator
- 14:50 A Compressed Encoding Scheme for Approximate TDOA Estimation
- 15:10 Robust Stochastic Maximum Likelihood Algorithm for DOA Estimation of Acoustic Sources in the Spherical Harmonic Domain
- 15:30 3D Localization of Multiple Simultaneous Speakers with Discrete Wavelet Transform and Proposed 3D Nested Microphone Array
- 15:50 Complexity-Reduced Solution for TDOA Source Localization in Large Equal Radius Scenario with Sensor Position Errors
SIPA-P1: Signal and Image Processing Applications
Chair: Patrick Le Callet (Université de Nantes, France)
- A Comparative Study of Orthogonal Moments for Micro-Doppler Classification
- Embedding Intelligent Features for Vibration-Based Machine Condition Monitoring
- Modelling Data with Both Sparsity and a Gaussian Random Field: Application to Dark Matter Mass Mapping in Cosmology
- Automated Tire Footprint Segmentation
- Advanced Cyclostationary-Based Analysis for Condition Monitoring of Complex Systems
- Speech Dereverberation Using Fully Convolutional Networks
SPCV-P1: Signal Processing for Computer Vision
Chair: Patrick Le Callet (Université de Nantes, France)
SPON-L1: Signal Processing over Networks
Room A
Chair: Paolo Di Lorenzo (Sapienza University of Rome, Italy)
- 14:30 Decisions Under Binary Messaging over Adaptive Networks
- 14:50 Efficient Variance-reduced Learning over Multi-agent Networks
- 15:10 Estimating the Topology of Neural Networks from Distributed Observations
- 15:30 Model-Based Voice Activity Detection in Wireless Acoustic Sensor Networks
- 15:50 Optimized Small Cell Range Expansion in Mobile Communication Networks Using Multi-Class Support Vector Machines
SPTM-P1: Signal Processing Theory and Methods
Chair: Stefania Colonnese (Università La Sapienza di Roma, Italy)
- Iterative Reconstruction of Spectrally Sparse Signals from Level Crossings
- Bayesian Inference with Error Variable Splitting and Sparsity Enforcing Priors for Linear Inverse Problems
- Sampling a Noisy Multiple Output Channel to Maximize the Capacity
- Roundoff Noise Analysis for Generalized Direct-Form II Structure of 2-D Separable-Denominator Digital Filters
- Low-Complexity RLS Algorithms for the Identification of Bilinear Forms
- Sequential Polynomial QR Decomposition and Decoding of Frequency Selective MIMO Channels
- A New Proportionate Adaptive Filtering Algorithm with Coefficient Reuse and Robustness Against Impulsive Noise
- Minimum Length Solution for One-Dimensional Discrete Phase Retrieval Problem
- Low-Rate Farrow Structure with Discrete-Lowpass and Polynomial Support for Audio Resampling
- Effect of Random Sampling on Noisy Nonsparse Signals in Time-Frequency Analysis
- Optimizing Approximate Message Passing for Variable Measurement Noise
- Heuristics for Efficient Sparse Blind Source Separation
- Improved Discrete Grey Wolf Optimizer
- Efficient Sampling Rate Offset Compensation - An Overlap-Save Based Approach
- Kernel Adaptive Hammerstein Filter
- Octonion Spectrum of 3D Octonion-Valued Signals - Properties and Possible Applications
SS-L3: Special Session on Security of Machine Learning for Virtual and Analog Worlds
Room Q
Chair: Battista Biggio (University of Cagliari & Pluribus One, Italy)
Organizers: Battista Biggio and Slava Voloshynovskiy
- 14:30 Effect of Training and Test Datasets on Image Restoration and Super-Resolution by Deep Learning
- 14:50 Adversarial Machine Learning Against Digital Watermarking
- 15:10 Explaining Black-box Android Malware Detection
- 15:30 Physical Layer Communications System Design Over-the-Air Using Adversarial Networks
- 15:50 Adversarial Malware Binaries: Evading Deep Learning for Malware Detection in Executables
SS-P1: Special Session on Biometric Authentication on Mobile Devices
Chair: Richard Guest (University of Kent, United Kingdom (Great Britain))
Organizers: Richard Guest and Christian Krätzer
- End-to-end Photoplethysmography (PPG) Based Biometric Authentication by Using Convolutional Neural Networks
- User Interaction in Mobile Biometrics
- A Requirement Analysis for Privacy Preserving Biometrics in View of Universal Human Rights and Data Protection Regulation
- A Framework for Assessing Factors Influencing User Interaction for Touch-based Biometrics
- Towards Understanding the Effects of Practice on Behavioural Biometric Recognition Performance
Tuesday, September 4 16:10 - 16:40
Communication Break
Tuesday, September 4 16:40 - 18:20
CS-L1: Compressed Sensing
Room G
Chair: Carlo S Regazzoni (University of Genoa, Italy)
- 16:40 Recovery of Linearly Mixed Sparse Sources from Multiple Measurement Vectors Using l1-Minimization
- 17:00 From L1 Minimization to Entropy Minimization: A Novel Approach for Sparse Signal Recovery in Compressive Sensing
- 17:20 Structured Dictionary Learning for Compressive Speech Sensing
- 17:40 Greedy Recovery of Sparse Signals with Dynamically Varying Support
- 18:00 Sensing Matrix Sensitivity to Random Gaussian Perturbations in Compressed Sensing
IVM-L1: Image, video and multimedia processing
Room: Auditorium
Chair: Domenico Vitulano (Consiglio Nazionale delle Ricerche, Italy)
- 16:40 A Colour Hit-or-Miss Transform Based on a Rank Ordered Distance Measure
- 17:00 Feature Trajectories Selection for Video Stabilization
- 17:20 Graph Spectral Domain Shape Representation
- 17:40 An Entropy-Based Approach for Shape Description
- 18:00 Octonion Sparse Representation for Color and Multispectral Image Processing
SENS-P1: Sensor Array and Multichannel Signal Processing
Chair: Maria S. Greco (University of Pisa, Italy)
- Coherence Constrained Alternating Least Squares
- A Fast Eigen-based Signal Combining Algorithm by Using CORDIC
- On the Angular Resolution Limit Uncertainty
- Asymmetric Supercardioid Beamforming Using Circular Microphone Arrays
- A New Beamformer Design Method for Multi-Group Multicasting by Enforcing Constructive Interference
- Time Modulated Array Controlled by Periodic Pulsed Signals
- Evolutionary Resampling for Multi-Target Tracking Using Probability Hypothesis Density Filter
- Superpixel Construction for Hyperspectral Unmixing
- Improved ADMM-Based Algorithm for Multi-Group Multicast Beamforming in Large-Scale Antenna Systems
- Free-walking 3D Pedestrian Large Trajectory Reconstruction from IMU Sensors
SIPA-P2: Signal and Image Processing Applications
Chair: Ido Nevat (TUMCREATE, Singapore)
- Optimal Estimation with Extended Battery Life in Wireless Sensor Networks
- An Efficient Machine Learning-Based Fall Detection Algorithm Using Local Binary Features
- Joint Localization and Clock Offset Estimation via Time-Of-Arrival with Ranging Offset
- CASSIS: Characterization with Adaptive Sample-Size Inferential Statistics Applied to Inexact Circuits
- Binarized Convolutional Neural Networks for Efficient Inference on GPUs
- Online Prediction of Robot to Human Handover Events Using Vibrations
- Fundamental Limits for Joint Relative Position and Orientation Estimation with Generic Antennas
- Visual Localization in the Presence of Appearance Changes Using the Partial Order Kernel
- Robotic Mobility Diversity Algorithm with Continuous Search Space
SPTM-L1: Signal Processing Theory and Methods
Room Pininfarina
Chair: Antonio Napolitano (Universita di Napoli Parthenope, Italy)
- 16:40 Data-Selective Conjugate Gradient Algorithm
- 17:00 Sparse Phase Retrieval via Iteratively Reweighted Amplitude Flow
- 17:20 Sparse Time-Frequency-Frequency-Rate Representation for Multicomponent Nonstationary Signal Analysis
- 17:40 On the Time-Frequency Reassignment of Interfering Modes in Multicomponent FM Signals
- 18:00 On Cyclostationarity-Based Signal Detection
SS-L4: Special Session on Advances in large-scale inverse problems in image processing
Room A
Chair: Miguel Simões (IST - IT, Portugal)
Organizers: Miguel Simões and Qi Wei
SS-L5: Special Session on SMARTHEART: Real-time technologies for cardiac signal processing
Room Q
Chair: Nuno Pombo (University of Beira Interior & BSAFE - Lab, Portugal)
Organizers: Nuno Pombo, Hugo Silva, and Miguel Coimbra
- 16:40 Convolutional Neural Networks for Heart Sound Segmentation
- 17:00 Estimation of Missing Data in Fetal Heart Rate Signals Using Shift-Invariant Dictionary
- 17:20 Sensitivity of the Contactless Videoplethysmography-Based Heart Rate Detection to Different Measurement Conditions
- 17:40 RepoBIT: Cloud-Driven Real-Time Biosignal Streaming, Storage, Visualisation and Sharing
- 18:00 An Efficient Lossless Compression Algorithm for Electrocardiogram Signals
SS-P2: Special Session on Cross-layer and joint network resources optimizations
Chair: Valerio Frascolla (Intel Deutschland Gmbh, Germany)
Organizer: Valerio Frascolla
- Dynamic Allocation of Processing Resources in Cloud-RAN for a Virtualised 5G Mobile Network
- Network Utility Maximization for Adaptive Resource Allocation in DSL Systems
- Proactive Computation Caching Policies for 5G-and-beyond Mobile Edge Cloud Networks
- Joint Optimization of Caching and Transport in Proactive Edge Cloud
- Cross-layer Optimization in Terminals
SS-P3: Special Session on Overcoming interference in next-generation wireless networks
Chair: Francesco Verde (University of Napoli Federico II & National Inter-University Consortium for Telecommunications, Italy)
Organizers: Francesco Verde and Symeon Chatzinotas
- Performance of a Third-Order VOLTERRA MVDR Beamformer in the Presence of non-Gaussian and/or Non-Circular Interference
- Hybrid Analog-Digital Precoding for Interference Exploitation
- Interference-based Clustering for MIMO D2D Underlay Communications
- Two-Step Hybrid Multiuser Equalizer for Sub-Connected mmWave Massive MIMO SC-FDMA Systems
- Sequential Spatio-temporal Symbol-level Precoding Enabling Faster-than-Nyquist Signaling for Multi-user MISO Systems
- How Much Will Tiny IoT Nodes Profit from Massive Base Station Arrays?
Wednesday, September 5 9:00 - 10:00
PL2: Synthetic Molecular Communication: Fundamentals, Opportunities, and Challenges
Robert Schober (Friedrich-Alexander University,Germany)
Room: AuditoriumWednesday, September 5 10:00 - 10:30
Communication Break
Wednesday, September 5 10:30 - 12:30
AASP-L2: Sound event detection and classification
Room Pininfarina
Chairs: Gyorgy Fazekas (Queen Mary University of London, United Kingdom (Great Britain)), Hélène Papadopoulos (CNRS, France)
- 10:30 A Layer-wise Score Level Ensemble Framework for Acoustic Scene Classification
- 10:50 Diarization and Separation Based on a Data-Driven Simplex
- 11:10 Modelling of Sound Events with Hidden Imbalances Based on Clustering and Separate Sub-Dictionary Learning
- 11:30 Connectionist Temporal Classification-based Sound Event Encoder for Converting Sound Events into Onomatopoeic Representations
- 11:50 Multi-Channel Non-Negative Matrix Factorization for Overlapped Acoustic Event Detection
- 12:10 Acoustic Scene Classification from Few Examples
GSP-P1: Graph Signal Processing
Chair: Paolo Di Lorenzo (Sapienza University of Rome, Italy)
- Graph-Based Inpainting of Disocclusion Holes for Zooming in 3D Scenes
- Improving Graph Convolutional Networks with Non-Parametric Activation Functions
- Graph Similarity Based on Graph Fourier Distances
- Graph Representation Using Mutual Information for Graph Model Discrimination
- Fusion of Community Structures in Multiplex Networks by Label Constraints
- Sampling and Reconstruction of Band-limited Graph Signals Using Graph Syndromes
- Rate Distortion Optimized Graph Partitioning for Omnidirectional Image Coding
- Estimation of Bandlimited Signals on Graphs from Single Bit Recordings of Noisy Samples
- Joint Graph Learning and Signal Recovery via Kalman Filter for Multivariate Autoregressive Processes
SPCOM-L1: Signal Processing for Communications
Room: Auditorium
Chair: Umberto Spagnolini (Politecnico di Milano, Italy)
- 10:30 Sampling Phase Estimation in Underwater PPM Fractionally Sampled Equalization
- 10:50 On Partial Response Signaling for MIMO Equalization on Multi-Gbit/s Electrical Interconnects
- 11:10 Low-Rank Channel and Interference Estimation in mm-Wave Massive Antenna Arrays
- 11:30 1-Bit Massive MIMO Downlink Based on Constructive Interference
- 11:50 Iterative Equalization Based on Expectation Propagation: a Frequency Domain Approach
- 12:10 Joint Long-Term Admission Control and Beamforming in Downlink MISO Networks
SS-L6: Special Session on Adversarial Multimedia Forensics
Room A
Chair: Benedetta Tondi (University of Siena, Italy)
Organizers: Mauro Barni, Matthew Stamm, and Benedetta Tondi
- 10:30 Rebroadcast Attacks: Defenses, Reattacks, and Redefenses
- 10:50 Detecting Adversarial Examples - A Lesson from Multimedia Security
- 11:10 Anti-forensics of JPEG Compression Using Generative Adversarial Networks
- 11:30 Fooling PRNU-Based Detectors Through Convolutional Neural Networks
- 11:50 Adversarial Multimedia Forensics: Overview and Challenges Ahead
- 12:10 Analysis of Adversarial Attacks Against CNN-based Image Forgery Detectors
SS-L7: Special Session on Computational Audio Processing in Real Acoustic Environments
Room Q
Chair: Michele Scarpiniti (Sapienza University of Rome, Italy)
Organizers: Michele Scarpiniti, Jen-Tzung Chien, and Stefano Squartini
- 10:30 Acoustic Event Classification Using Multi-resolution HMM
- 10:50 A Fusion of Deep Convolutional Generative Adversarial Networks and Sequence to Sequence Autoencoders for Acoustic Scene Classification
- 11:10 On Room Impulse Response Measurement Using Perfect Sequences for Wiener Nonlinear Filters
- 11:30 Beamforming-Based Acoustic Source Localization and Enhancement for Multirotor UAVs
- 11:50 Infant Cry Detection in Adverse Acoustic Environments by Using Deep Neural Networks
- 12:10 A CNN-GRU Approach to Capture Time-Frequency Pattern Interdependence for Snore Sound Classification
SS-L8: Special Session on Morphing detection in biometric images
Room G
Chair: Andrey Makrushin (University of Magdeburg, Germany)
Organizers: Andreas Wolf and Andrey Makrushin
- 10:30 Reducing the False Alarm Rate for Face Morph Detection by a Morph Pipeline Footprint Detector
- 10:50 Automatic Detection of Image Morphing by Topology-based Analysis
- 11:10 Morphing Detection Using a General-Purpose Face Recognition System
- 11:30 An Overview of Recent Advances in Assessing and Mitigating the Face Morphing Attack
- 11:50 Reflection Analysis for Face Morphing Attack Detection
- 12:10 Towards Robust Evaluation of Face Morphing Detection
SS-P4: Special Session on Recent Advances in Advanced Driver Assistance using Signal Processing, Computer Vision and AI
Chair: Marcos Nieto (Vicomtech, Spain)
Organizers: Marcos Nieto, Suzanne Little, Oihana Otaegui, and Noel E. O'Connor
- Fast Multi-Lane Detection and Modeling for Embedded Platforms
- The Role of Cloud-Computing in the Development and Application of ADAS
- Bayesian Track-to-Graph Association for Maritime Traffic Monitoring
- Autonomous Person Detection and Tracking Framework Using Unmanned Aerial Vehicles (UAVs)
- A Residual Encoder-Decoder Network for Semantic Segmentation in Autonomous Driving Scenarios
SS-P5: Special Session on Representation Learning in Signal Processing
Chairs: Emilie Chouzenoux (Université Paris-Est Marne-la-Vallée, France), Angshul Majumdar (Indraprastha Institute Of Information Technology-Delhi & University of British Columbia, India)
Organizers: Emilie Chouzenoux and Angshul Majumdar
- Transformed Locally Linear Manifold Clustering
- Jacobi Algorithm for Nonnegative Matrix Factorization with Transform Learning
- Coupled Autoencoder Based Reconstruction of Images from Compressively Sampled Measurements
- Graph Structured Dictionary for Regression
- Sparse Multiple Kernel Learning: Support Identification via Mirror Stratifiability
SSP-P1: Statistical Signal Processing
Chair: Sergio Barbarossa (Sapienza University of Rome, Italy)
- Complete Model Selection in Multiset Canonical Correlation Analysis
- Theoretical Study of Multiscale Permutation Entropy on Finite-length Fractional Gaussian Noise
- Emotion Estimation in Crowds: The Interplay of Motivations and Expectations in Individual Emotions
- Robust Detection and Estimation of Change-Points in a Time Series of Multivariate Images
- On the Detection of Low Rank Matrices in the High-Dimensional Regime
- Improving Portfolios Global Performance with Robust Covariance Matrix Estimation: Application to the Maximum Variety Portfolio
- Range Estimation from Single-Photon Lidar Data Using a Stochastic EM Approach
- Modeling Time of Arrival Probability Distribution and TDOA Bias in Acoustic Emission Testing
- A Factorization Approach to Smoothing of Hidden Reciprocal Models
- Multi-Bernoulli Filter Implementation in Images
Wednesday, September 5 12:30 - 14:00
Lunch
Wednesday, September 5 12:30 - 13:00
3MT: 3 Minute Thesis
Room: Auditorium
Wednesday, September 5 14:00 - 15:40
BSIP+BINSP-P2: Biomedical signal and image processing & Bio-inspired signal and image processing and genomics
Chair: Gloria Menegaz (University of Verona, Italy)
- Computational Diagnosis of Parkinson's Disease from Speech Based on Regularization Methods
- Obstructive Sleep Apnea (OSA) Classification Using Analysis of Breathing Sounds During Speech
- Pipeline Comparison for the Pre-Processing of Resting-State Data in Epilepsy
- Classification Between Abnormal and Normal Respiration Through Observation Rate of Heart Sounds Within Lung Sounds
- Assessing Tissue Heterogeneity by non-Gaussian Measures in a Permeable Environment
- Recurrent Neural Networks with Weighting Loss for Early Prediction of Hand Movements
- Deep Neuroevolution: Training Deep Neural Networks for False Alarm Detection in Intensive Care Units
- A Data Driven Empirical Iterative Algorithm for GSR Signal Pre-Processing
- Improved Protein Residue-Residue Contact Prediction Using Image Denoising Methods
- Design of Optimal Frequency-Selective FIR Filters Using a Memetic Algorithm
- Metagenomic Composition Analysis of Sedimentary Ancient DNA from the Isle of Wight
ML-L1: Machine Learning
Room: Auditorium
Chair: Carlos J. Pérez (University of Extremadura, Spain)
- 14:00 A New Asymmetric Link-Based Binary Regression Model to Detect Parkinson's Disease by Using Replicated Voice Recordings
- 14:20 Spectral Clustering with Automatic Cluster-Number Identification via Finding Sparse Eigenvectors
- 14:40 Semi-supervised Learning for Dynamic Modeling of Brain Signals During Visual and Auditory Tests
- 15:00 Active Learning for One-Class Classification Using Two One-Class Classifiers
- 15:20 Multiscale DCNN Ensemble Applied to Human Activity Recognition Based on Wearable Sensors
RADAR-L1: Remote Sensing
Room G
Chair: Alfonso Farina (Leonardo Company Consultant, Italy)
- 14:00 New Doppler Processing for the Detection of Small and Slowly-Moving Targets in Highly Ambiguous Radar Context
- 14:20 Resolution Enhancement and Interference Suppression for Planetary Radar Sounders
- 14:40 ECA Filter Effects on Ground Clutter Statistics in DVB-T Based Passive Radar
- 15:00 Efficient Inverse Scattering Algorithm by Incorporating RPM Method for Microwave Non-destructive Imaging
- 15:20 Short-Duration Doppler Spectrogram for Person Recognition with a Handheld Radar
SENS-L2: Sensor Arrays: Channel Identification and Source Separation
Room Pininfarina
Chair: Shoji Makino (University of Tsukuba, Japan)
- 14:00 Nonunitary Joint Diagonalization for Overdetermined Convolutive Blind Signal Separation
- 14:20 Direction-of-Arrival Estimation for Uniform Rectangular Array: A Multilinear Projection Approach
- 14:40 A Bayesian Blind Source Separation Method for a Linear-quadratic Model
- 15:00 Efficient and Stable Joint Eigenvalue Decomposition Based on Generalized Givens Rotations
- 15:20 Identification of Multichannel AR Models with Additive Noise: a Frisch Scheme Approach
SPCOM-P1: Signal Processing for Communications
Chair: Paolo Di Lorenzo (Sapienza University of Rome, Italy)
- Arbitrary Length Perfect Integer Sequences Using Geometric Series
- Decision Statistics for Noncoherent Signal Detection in Multi-Element Antenna Arrays
- Performance Bounds Analysis for Semi-Blind Channel Estimation with Pilot Contamination in Massive MIMO-OFDM Systems
- Performance Analysis of Uplink Massive MIMO System over Rician Fading Channel
- Optimal SWIPT Beamforming for MISO Interfering Broadcast Channels with Multi-Type Receivers
- Efficient Semi-Blind Subspace Channel Estimation for MIMO-OFDM System
- A Frequency-Domain Band-MMSE Equalizer for Continuous Phase Modulation over Frequency-Selective Time-Varying Channels
- DQLC Optimization for Joint Source Channel Coding of Correlated Sources over Fading MAC
- Channel Estimation Based on the Discrete Cosine Transform Type-III Even
- An Enhanced Interleaving Frame Loss Concealment Method for Voice over IP Network Services
SPCV-L1: Signal Processing for Computer Vision
Room A
Chair: Giuseppe Valenzise (CNRS, Italy)
- 14:00 Characterizing 3D Shapes: a Complex Network-Based Approach
- 14:20 Texture Classification Using Fractal Dimension Improved by Local Binary Patterns
- 14:40 Convolutional Neural Networks Without Any Checkerboard Artifacts
- 15:00 A Holistic Automatic Method for Grouping Facial Features Based on Their Appearance
- 15:20 Compound Facial Expressions in 3D Faces: Forced and Spontaneous
SPTM-L2: Signal Processing Theory and Methods
Room Q
Chair: Abdelhak M Zoubir (Darmstadt University of Technology, Germany)
- 14:00 Space Alternating Variational Bayesian Learning for LMMSE Filtering
- 14:20 New Results on LMVDR Estimators for LDSS Models
- 14:40 Numerical Stability of Spline-Based Gabor-like Systems
- 15:00 Bayesian Target Enumeration and Labeling Using Radar Data of Human Gait
- 15:20 Multifractal Characterization for Bivariate Data
SS-P6: Special Session on What's new with the PRNU?
Chair: Fernando Pérez-González (University of Vigo, Spain)
Organizer: Fernando Pérez-González
- Putting the PRNU Model in Reverse Gear: Findings with Synthetic Signals
- PRNU-based Image Classification of Origin Social Network with CNN
- Blind Detection and Localization of Video Temporal Splicing Exploiting Sensor-Based Footprints
- Extracting PRNU Noise from H.264 Coded Videos
- Camera-based Image Forgery Localization Using Convolutional Neural Networks
Wednesday, September 5 15:40 - 16:10
Communication Break
Wednesday, September 5 16:10 - 17:50
BSIP-L2: Biomedical signal and image processing
Room Pininfarina
Chair: Philippe Ciuciu (CEA/NeuroSpin & INRIA Saclay, France)
- 16:10 Characterization of Mental States Through Node Connectivity Between Brain Signals
- 16:30 On the Most Informative Slice of Bicoherence That Characterizes Resting State Brain Connectivity
- 16:50 An Empirical Evaluation of Short-Term Memory Retention Using Different High-density EEG Based Brain Connectivity Measures
- 17:10 Connectivity Modulations Induced by Reaching&Grasping Movements
- 17:30 Scale-free Functional Connectivity Analysis from Source Reconstructed MEG Data
IFS-P1: Information forensics and security
Chair: Dinu Coltuc ("Valahia" University of Targoviste, Romania)
- P-score: Performance Aligned Normalization and an Evaluation in Score-level Multi-biometric Fusion
- Ear Acoustic Biometrics Using Inaudible Signals and Its Application to Continuous User Authentication
- Improved Pairwise Embedding for High-Fidelity Reversible Data Hiding
- Active Content Fingerprinting Using Latent Data Representation, Extractor and Reconstructor
- Quantitative and Binary Steganalysis in JPEG: A Comparative Study
- Phylogenetic Analysis of Multimedia Codec Software
- Non-intrusive Fingerprints Extraction from Hyperspectral Imagery
IVM-L2: Image, video and multimedia processing
Room: Auditorium
Chair: Yuk Hee Chan (The Hong Kong Polytechnic University, Hong Kong)
- 16:10 Performance Evaluation of No-Reference Image Quality Metrics for Visible Wavelength Iris Biometric Images
- 16:30 Impacts of Viewing Conditions on HDR-VDP2
- 16:50 Information Fusion Based Quality Enhancement for 3D Stereo Images Using CNN
- 17:10 A Blind Image Quality Metric Using a Selection of Relevant Patches Based on Convolutional Neural Network
- 17:30 Colorblind-friendly Halftoning
ML-P1: Machine Learning
Chair: Pedro Comesaña (University of Vigo, Spain)
- Direction of Arrival Estimation for Multiple Sound Sources Using Convolutional Recurrent Neural Network
- Deep Convolutional Neural Networks for Chaos Identification in Signal Processing
- Elastic Neural Networks: A Scalable Framework for Embedded Computer Vision
- Bilinear Residual Neural Network for the Identification and Forecasting of Geophysical Dynamics
- Moving Target Classification in Automotive Radar Systems Using Convolutional Recurrent Neural Networks
- Optimized Binary Hashing Codes Generated by Siamese Neural Networks for Image Retrieval
- Real-Time Hand Gesture Recognition Based on Artificial Feed-Forward Neural Networks and EMG
- Hierarchic ConvNets Framework for Rare Sound Event Detection
SS-L10: Special Session on System Level Optimization and Evaluation of 5G and Beyond Mobile Communications
Room G
Chair: Martin Klaus Müller (TU Wien, Austria)
Organizers: Martin Müller, Markus Rupp, and Stefan Schwarz
- 16:10 Resource Allocation for QF VMIMO Receive Cooperation in Urban Traffic Hotspots
- 16:30 Stochastic Geometry Modeling of Cellular Networks: A New Definition of Coverage and Its Application to Energy Efficiency Optimization
- 16:50 Effective Network Area for Efficient Simulation of Finite Area Wireless Networks
- 17:10 Sparse Beamforming for mmWave Spectrum Sharing Systems
- 17:30 Coverage-Improvement of V2I Communication Through Car-Relays in Microcellular Urban Networks
Wednesday, September 5 16:10 - 18:10
SS-L9: Special Session on Distant speech processing for smart speakers
Room Q
Chairs: Daniele Giacobello (Sonos, USA), Romain Serizel (Université de Lorraine, LORIA, France)
Organizers: Daniele Giacobello, Francesco Nesta, and Romain Serizel
- 16:10 Multichannel Audio Front-End for Far-Field Automatic Speech Recognition
- 16:30 Collaborative Speech Dereverberation: Regularized Tensor Factorization for Crowdsourced Multi-channel Recordings
- 16:50 Reconstruction of the Virtual Microphone Signal Based on the Distributed Ray Space Transform
- 17:10 Recursive Expectation-Maximization Algorithm for Online Multi-Microphone Noise Reduction
- 17:30 An Online Expectation-Maximization Algorithm for Tracking Acoustic Sources in Multi-Microphone Devices During Music Playback
- 17:50 Neural-Network Supervised Maximum Likelihood-based On-Line Dereverberation
Wednesday, September 5 16:10 - 17:50
SS-P7: Special Session on Acoustic Scene Analysis and Signal Enhancement Based on Advanced Signal Processing and Machine Learning
Chairs: Shoji Makino (University of Tsukuba, Japan), Nobutaka Ono (Tokyo Metropolitan University, Japan)
Organizers: Shoji Makino and Nobutaka Ono
- Independent Deeply Learned Matrix Analysis for Multichannel Audio Source Separation
- LCMV Beamformer with DNN-based Multichannel Concurrent Speakers Detector
- Deep Neural Networks for Joint Voice Activity Detection and Speaker Localization
- Learning-based Acoustic Source Localization in Acoustic Sensor Networks Using the Coherent-to-Diffuse Power Ratio
- Raw Multi-Channel Audio Source Separation Using Multi-Resolution Convolutional Auto-Encoders
- Time-Frequency-Bin-Wise Beamformer Selection and Masking for Speech Enhancement in Underdetermined Noisy Scenarios
SS-P8: Special Session on Emerging Data Structure Paradigms for Subspace Estimation
Chairs: Arnaud Breloy (University Paris Nanterre, France), Mohammed Nabil El Korso (Paris 10 University & LEME-EA 4416, France)
Organizers: Arnaud Breloy, Mohammed Nabil El Korso, and Hamid Krim
- Parallel and Hybrid Soft-Thresholding Algorithms with Line Search for Sparse Nonlinear Regression
- Simultaneous Signal Subspace Rank and Model Selection with an Application to Single-Snapshot Source Localization
- Learning Dictionary-Based Unions of Subspaces for Image Denoising
- Robust Subspace Clustering for Radar Detection
- Small Variance Asymptotics and Bayesian Nonparametrics for Dictionary Learning
- Information Subspace-Based Fusion for Vehicle Classification
SSP-L2: Statistical Signal Processing
Room A
Chair: Vincenzo Matta (University of Salerno, Italy)
- 16:10 Tracking and Sensor Fusion in Direction of Arrival Estimation Using Optimal Mass Transport
- 16:30 Robustifying Sequential Multiple Hypothesis Tests in Distributed Sensor Networks
- 16:50 Prior Influence on Weiss-Weinstein Bounds for Multiple Change-Point Estimation
- 17:10 Improved Adaptive Importance Sampling Based on Variational Inference
- 17:30 In Search for Improved Auxiliary Particle Filters
Wednesday, September 5 20:00 - 23:00
Gala Dinner
Thursday, September 6 9:00 - 10:00
PL3: Deep Convolutional Networks: An Opportunity for Signal Processing
Stéphane Mallat (Collège de France, France)
Room: AuditoriumThursday, September 6 10:00 - 10:30
Communication Break
Thursday, September 6 10:30 - 12:30
AASP-L3: Audio and speech source separation
Room: Auditorium
Chair: Patrick A Naylor (Imperial College London, United Kingdom (Great Britain))
- 10:30 Speech Enhancement Using Kalman Filtering in the Logarithmic Bark Power Spectral Domain
- 10:50 Enhanced Time-Frequency Masking by Using Neural Networks for Monaural Source Separation in Reverberant Room Environments
- 11:10 Independent Positive Semidefinite Tensor Analysis in Blind Source Separation
- 11:30 Independent Low-Rank Tensor Analysis for Audio Source Separation
- 11:50 Noisy cGMM: Complex Gaussian Mixture Model with Non-Sparse Noise Model for Joint Source Separation and Denoising
- 12:10 FastFCA: Joint Diagonalization Based Acceleration of Audio Source Separation Using a Full-Rank Spatial Covariance Model
BSIP-L3: Biomedical signal and image processing
Room G
Chair: Puthusserypady Sadasivan (Technical University of Denmark, Denmark)
- 10:30 Modeling the Visual Pathway for Stimulus Optimization in Brain-Computer Interfaces
- 10:50 A Deep Learning MI-EEG Classification Model for BCIs
- 11:10 Performance of Nested vs. Non-Nested SVM Cross-Validation Methods in Visual BCI: Validation Study
- 11:30 Transfer Learning for SSVEP-based BCI Using Riemannian Similarities Between Users
- 11:50 Fast and Accurate Multiclass Inference for MI-BCIs Using Large Multiscale Temporal and Spectral Features
- 12:10 Use of Topological Data Analysis in Motor Intention Based Brain-Computer Interfaces
CS-P1: Compressed Sensing
Chair: Michele Scarpiniti (Sapienza University of Rome, Italy)
- Defect Detection from 3D Ultrasonic Measurements Using Matrix-free Sparse Recovery Algorithms
- Compressed Edge Spectrum Sensing for Wideband Cognitive Radios
- Compressive Sensing of Temporally Correlated Sources Using Isotropic Multivariate Stable Laws
- A Proximal Method for Convolutional Dictionary Learning with Convergence Property
- Approximate Recovery of Initial Point-like and Instantaneous Sources from Coarsely Sampled Thermal Fields via Infinite-Dimensional Compressed Sensing
- Joint Low Mutual and Average Coherence Dictionary Learning
- Sparse Method for Tip-Timing Signals Analysis with Non Stationary Engine Rotation Frequency
- Injecting Image Priors into Learnable Compressive Subsampling
- Dictionary Learning for Photometric Redshift Estimation
- Computationally Efficient Estimation of Multi-dimensional Damped Modes Using Sparse Wideband Dictionaries
- Structured Sparsity Regularization for Gravitational-Wave Polarization Reconstruction
- Sparse Time-Frequency Representation of Gravitational-Wave Signals in Unions of Wilson Bases
- Robust Data Hiding Scheme for Compressively Sensed Signals
- Rethinking Compressive Sensing
- Incoherent Projection Matrix Design for Compressed Sensing Using Alternating Optimization
IVM-P2: Image, video and multimedia processing
Chair: Nikos Nikolaidis (Aristotle University of Thessaloniki, Greece)
- Real-Time Deep Learning Method for Abandoned Luggage Detection in Video
- Respiratory Rate Monitoring by Video Processing Using Local Motion Magnification
- A Vector of Locally Aggregated Descriptors Framework for Action Recognition on Motion Capture Data
- Image Analysis Based Fish Tail Beat Frequency Estimation for Fishway Efficiency
- Face Frontalization for Cross-Pose Facial Expression Recognition
- Inferring User Gender from User Generated Visual Content on a Deep Semantic Space
- Towards Automatic Detection of Animals in Camera-Trap Images
SPCOM-L2: Signal Processing for Communications
Room Pininfarina
Chair: Emilio Calvanese Strinati (CEA-LETI, France)
- 10:30 Extrapolated Projection Methods for PAPR Reduction
- 10:50 Energy Balancing for Robotic Aided Clustered Wireless Sensor Networks Using Mobility Diversity Algorithms
- 11:10 Perturbation Analysis of Root-MUSIC-Type Methods for Blind Network-Assisted Diversity Multiple Access
- 11:30 Efficient Channel Estimation in Millimeter Wave Hybrid MIMO Systems with Low Resolution ADCs
- 11:50 Dedicated Beam-Based Channel Training Technique for Millimeter Wave Communications with High Mobility
- 12:10 Blind Spectrum Sensing Based on Recurrence Quantification Analysis in the Context of Cognitive Radio
SS-L11: Special Session on Light Field Coding
Room Q
Chair: Luis Ducla Soares (I.S.C.T.E. / I.T. - Lisbon, Portugal)
Organizers: Luís Ducla Soares and Paulo Nunes
- 10:30 Efficient Light Field Image Coding with Depth Estimation and View Synthesis
- 10:50 Light Field Image Coding Using High Order Prediction Training
- 11:10 Rate-Distortion Optimized Super-Ray Merging for Light Field Compression
- 11:30 Scalable Light Field Coding with Support for Region of Interest Enhancement
- 11:50 Light Field Compression of HDCA Images Combining Linear Prediction and JPEG 2000
- 12:10 Comparison of Interactive Subjective Methodologies for Light Field Quality Evaluation
SS-L12: Special Session on Understanding and design of deep neural networks
Room A
Chairs: Saikat Chatterjee (KTH - Royal Institute of Technology & Communication Theory Lab, Sweden), Nikos Deligiannis (Vrije Universiteit Brussel, Belgium), Pascal Frossard (EPFL, Switzerland)
Organizers: Nikos Deligiannis, Saikat Chatterjee, and Pascal Frossard
- 10:30 A Comparison of Audio Signal Preprocessing Methods for Deep Neural Networks on Music Tagging
- 10:50 Classification Asymptotics in the Random Matrix Regime
- 11:10 Regularizing Autoencoder-Based Matrix Completion Models via Manifold Learning
- 11:30 Task-Driven Dictionary Learning Based on Convolutional Neural Network Features
- 11:50 Fast Convergence for Stochastic and Distributed Gradient Descent in the Interpolation Limit
- 12:10 On Deep Learning for Inverse Problems
Thursday, September 6 12:30 - 14:00
Lunch
Thursday, September 6 14:00 - 15:40
AASP-L4: Music signal modeling, analysis, processing and synthesis
Room A
Chair: Toon van Waterschoot (KU Leuven, Belgium)
- 14:00 Automatic Chord Recognition with Higher-Order Harmonic Language Modelling
- 14:20 Convolving Gaussian Kernels for RNN-based Beat Tracking
- 14:40 A Hierarchical Latent Mixture Model for Polyphonic Music Analysis
- 15:00 Beat Tracking Using Recurrent Neural Network: a Transfer Learning Approach
- 15:20 Unsupervised Singing Voice Separation Based on Robust Principal Component Analysis Exploiting Rank-1 Constraint
BSIP-L4: Biomedical signal and image processing
Room: Auditorium
Chair: Dimitri Van De Ville (Ecole Polytechnique Fédérale de Lausanne, Switzerland)
- 14:00 Real-time DCT Learning-based Reconstruction of Neural Signals
- 14:20 Combined Analysis-l1 and Total Variation ADMM with Applications to MEG Brain Imaging and Signal Reconstruction
- 14:40 Improving EEG Source Localization Through Spatio-Temporal Sparse Bayesian Learning
- 15:00 Analysis of Activity States of Local Neuronal Microcircuits in Mouse Brain
- 15:20 A New Approach to Sample Entropy of Multi-channel Signals: Application to EEG Signals
CS-L2: Compressed Sensing
Room Pininfarina
Chair: Ana Isabel Pérez-Neira (Universitat Politècnica de Catalunya, Spain)
- 14:00 Simultaneous Estimation of a System Matrix by Compressed Sensing and Finding Optimal Regularization Parameters for the Inversion Problem
- 14:20 Path Orthogonal Matching Pursuit for k-Sparse Image Reconstruction
- 14:40 Object Detection on Compressive Measurements Using Correlation Filters and Sparse Representation
- 15:00 Knowledge-Aided Normalized Iterative Hard Thresholding Algorithms for Sparse Recovery
- 15:20 Signal Reconstruction from Sub-sampled and Nonlinearly Distorted Observations
CSI-L1: Computational and Spectral Imaging
Room G
Chair: Henry Arguello (Universidad Industrial de Santander, Colombia)
- 14:00 On the Impact of Tensor Completion in the Classification of Undersampled Hyperspectral Imagery
- 14:20 Spatio-Spectral Multichannel Reconstruction from Few Low-Resolution Multispectral Data
- 14:40 Spectral Image Fusion from Compressive Projections Using Total-Variation and Low-Rank Regularizations
- 15:00 Graph Manifold Clustering Based Band Selection for Hyperspectral Face Recognition
- 15:20 Multi-Resolution Reconstructions from Compressive Spectral Coded Projections
ML-P2: Machine Learning
Chair: Hiroyuki Kasai (The University of Electro-Communications, Japan)
- Sparse Autoencoders Using Non-smooth Regularization
- Classification by Re-Generation: Towards Classification Based on Variational Inference
- Riemannian Joint Dimensionality Reduction and Dictionary Learning on Symmetric Positive Definite Manifolds
- Trust-Region Minimization Algorithm for Training Responses (TRMinATR): The Rise of Machine Learning Techniques
- Novel Algorithm for Incremental L1-Norm Principal-Component Analysis
- Blind Multi-class Ensemble Learning with Dependent Classifiers
- A Deep Reinforcement Learning Approach for Early Classification of Time Series
- Automated Detection of Solar Cell Defects with Deep Learning
- Machine Learning for User Traffic Classification in Wireless Systems
- Skeleton-based Action Recognition Based on Deep Learning and Grassmannian Pyramids
- Machine Learning Based Indoor Localization Using a Representative k-Nearest-Neighbor Classifier on a Low-Cost IoT-Hardware
- An Unsupervised Frame Selection Technique for Robust Emotion Recognition in Noisy Speech
- Abnormal Behavior Detection in Crowded Scenes Using Density Heatmaps and Optical Flow
SLP-P1: Speech Processing and Human Language Technology
Chair: Gerald Enzner (Ruhr-Universität Bochum, Germany)
- Dual-Channel VTS Feature Compensation with Improved Posterior Estimation
- Speaking Rate Changes Affect Phone Durations Differently for Neutral and Emotional Speech
- Estimation of Pitch Targets from Speech Signals by Joint Regularized Optimization
- Hypo and Hyperarticulated Speech Data Augmentation for Spontaneous Speech Recognition
- Nonlinear Prediction of Speech by Echo State Networks
- Application-Layer Redundancy for the EVS Codec
- Reduced-Complexity Semi-Distributed Multi-Channel Multi-Frame MVDR Filter
- CycleGAN-VC: Non-parallel Voice Conversion Using Cycle-Consistent Adversarial Networks
- On the Automatic Validation of Speech Alignment
- Centerline Articulatory Models of the Velum and Epiglottis for Articulatory Synthesis of Speech
- Electrolaryngeal Speech Enhancement with Statistical Voice Conversion Based on CLDNN
SS-L13: Special Session on On the Suitability of Signal Processing Strategies for the Assessment of Road Pavement Surface Condition
Room Q
Chair: Henrique Oliveira (Instituto de Telecomunicações & Instituto Politécnico de Beja, Portugal)
Organizers: James Tsai and Henrique Oliveira
- 14:00 A Deep Convolutional Neural Network for Semantic Pixel-Wise Segmentation of Road and Pavement Surface Cracks
- 14:20 Municipal Infrastructure Anomaly and Defect Detection
- 14:40 An Acoustic Image-Source Characterisation of Surface Profiles
- 15:00 Road Surface Crack Detection Using a Light Field Camera
- 15:20 A Fast and Accurate Automated Pavement Crack Detection Algorithm
SS-P9: Special Session on Novel trends in signal processing towards user immersivity
Chair: Federica Battisti (Università degli Studi Roma Tre, Italy)
Organizers: Francesco De Natale and Federica Battisti
- Unsupervised Calibration of RGB-NIR Capture Pairs Utilizing Dense Multimodal Image Correspondences
- Lenslet Light Field Imaging Scalable Coding
- A Study on the Impact of Visualization Techniques on Light Field Perception
- Real-Time Quality Assessment of Videos from Body-Worn Cameras
- Intrinsic Light Field Decomposition and Disparity Estimation with Deep Encoder-Decoder Network
- Virtual Camera Modeling for Multi-View Simulation of Surveillance Scenes
Thursday, September 6 15:40 - 16:10
Communication Break
Thursday, September 6 16:10 - 17:50
AASP-L5: Modeling, Analysis and Synthesis of Acoustic Environments
Room: Auditorium
Chair: Roland Badeau (LTCI, Télécom ParisTech, Université Paris-Saclay, France)
- 16:10 Unified Stochastic Reverberation Modeling
- 16:30 Simultaneous Measurement of Spatial Room Impulse Responses from Multiple Sound Sources Using a Continuously Moving Microphone
- 16:50 Audio Virtualization of Façade Acoustic Insulation by Convex Optimization
- 17:10 Sparsity Based Framework for Spatial Sound Reproduction in Spherical Harmonic Domain
- 17:30 Efficient ML-Estimator for Blind Reverberation Time Estimation
CSI-P1: Computational and Spectral Imaging
Chair: Karen Egiazarian (Tampere University of Technology, Finland)
- Multi-frequency Phase Retrieval from Noisy Data
- Plenoptic Sensor: Application to Extend Field-Of-View
- A Fast Endmember Estimation Algorithm from Compressive Measurements
- Multi-Resolution Coded Apertures Based on Side Information for Single Pixel Spectral Reconstruction
- A Trace Lasso Regularized L1-norm Graph Cut for Highly Correlated Noisy Hyperspectral Image
- Fast Hyperspectral Cube Reconstruction for a Double Disperser Imager
IVM-L3: Image, video and multimedia processing
Room A
Chair: Achanna Anil Kumar (TCS Innovation Labs, India)
- 16:10 Poisson Image Denoising Using Best Linear Prediction: A Post-processing Framework
- 16:30 Satellite Image Segmentation with Deep Residual Architectures for Time-Critical Applications
- 16:50 Adaptive Guided Upsampling for Color Image Demosaicking
- 17:10 Computationally Efficient Image Super Resolution from Totally Aliased Low Resolution Images
- 17:30 On Multi-View Face Recognition Using Lytro Images
ML-L2: Machine Learning
Room Pininfarina
Chair: Christoph Studer (Cornell University, USA)
- 16:10 Capsule Routing for Sound Event Detection
- 16:30 "What are You Listening to?" Explaining Predictions of Deep Machine Listening Systems
- 16:50 Comparative Study on Spoken Language Identification Based on Deep Learning
- 17:10 Playlist-Based Tag Propagation for Improving Music Auto-Tagging
- 17:30 Analog-to-Feature (A2F) Conversion for Audio-Event Classification
RADAR-P1: Remote Sensing
Chair: Danilo Comminiello (Sapienza University of Rome, Italy)
- Multi-Satellite Cycle-Slip Detection and Exclusion Using the Noise Subspace of Residual Dynamics
- Optimum Detection for a Class of Stationary Meteorological Radars
- Classification of Intra-Pulse Modulation of Radar Signals by Feature Fusion Based Convolutional Neural Networks
- Classification of Volcano-Seismic Signals with Bayesian Neural Networks
- Decentralized Sensor Localization by Decision Fusion of RSSI and Mobility in Indoor Environments
SLP-L1: Speech Processing and Human Language Technology
Room G
Chair: Mads Græsbøll Christensen (Aalborg University, Denmark)
- 16:10 Integrating Denoising Autoencoder and Vector Taylor Series with Auditory Masking for Speech Recognition in Noisy Conditions
- 16:30 Automatic Speech Pronunciation Correction with Dynamic Frequency Warping-Based Spectral Conversion
- 16:50 Automatic Speech Translation System Selecting Target Language by Direction-of-Arrival Information
- 17:10 Online Parametric NMF for Speech Enhancement
- 17:30 On Optimal Filtering for Speech Decomposition
SPTM-L3: Signal Processing Theory and Methods
Room Q
Chair: Marius Pesavento (Technische Universität Darmstadt & Merckstr. 25, Germany)
- 16:10 Estimating Faults Modes in Ball Bearing Machinery Using a Sparse Reconstruction Framework
- 16:30 A Sequence-Filter Joint Optimization
- 16:50 The Matched Window Reassignment
- 17:10 Least Squares and Maximum Likelihood Estimation of Mixed Spectra
- 17:30 A Novel Method for Topological Embedding of Time-Series Data
SS-P10: Special Session on Security and Privacy in Biometrics
Chair: Paulo Lobato Correia (Instituto Superior Tecnico - Universidade Tecnica Lisboa & Instituto de Telecomunicacoes, Portugal)
Organizers: Christian Rathgeb, Marta Gomez-Barrero, and Javier Galbally
- Ear Presentation Attack Detection: Benchmarking Study with First Lenslet Light Field Database
- Privacy-Preserving Indexing of Iris-Codes with Cancelable Bloom Filter-based Search Structures
- Face Demorphing in the Presence of Facial Appearance Variations
- Fingerprint Minutiae Matching Through Sparse Cross-correlation
- Speaker Inconsistency Detection in Tampered Video
SSA-P1: Signal Processing Applications
Chair: Michele Scarpiniti (Sapienza University of Rome, Italy)
- Comparative Study on Univariate Forecasting Methods for Meteorological Time Series
- A New Blind Beamforming Technique for the Alignment and Enhancement of Seismic Signals
- An Improved CSI Based Device Free Indoor Localization Using Machine Learning Based Classification Approach
- Binary Sequences Set with Small ISL for MIMO Radar Systems
- DC-offset Estimation of Multiple CW Micro Doppler Radar
- Separation of Anthropogenic Noise and Extremely Low Frequency Natural Magnetic Field Using Statistical Features
- Mosquito Wingbeat Analysis and Classification Using Deep Learning
- Parameters Estimation of Ultrasonics Echoes Using the Cuckoo Search and Adaptive Cuckoo Search Algorithms
Friday, September 7 9:00 - 10:00
PL4: Sensing and Processing with Events
Tobi Delbruck (University of Zurich and ETH Zurich, Switzerland)
Room: AuditoriumFriday, September 7 10:00 - 10:30
Communication Break
Friday, September 7 10:30 - 12:30
AASP-P1: Audio and Acoustic Signal Processing
Chair: Mark Sandler (Queen Mary University of London, United Kingdom (Great Britain))
- Ultrasonic Based Proximity Detection for Handsets
- Improved Direct-path Dominance Test for Speaker Localization in Reverberant Environments
- Evaluation of Binaural Noise Reduction Methods in Terms of Intelligibility and Perceived Localization
- Estimating Power Spectral Density of Unmanned Aerial Vehicle Rotor Noise Using Multisensory Information
- Acoustic Scene Analysis Using Partially Connected Microphones Based on Graph Cepstrum
- Convolutional Recurrent Neural Networks for Urban Sound Classification Using Raw Waveforms
- Dynamics and Periodicity Based Multirate Fast Transient-Sound Detection
- A Convex-Combined Step-Size-Based Normalized Modified Filtered-x Least Mean Square Algorithm for Impulsive Active Noise Control Systems
- Lossy Audio Compression Identification
- Missing Sample Estimation Based on High-Order Sparse Linear Prediction for Audio Signals
- Parameter Domain Loudness Estimation in Parametric Audio Object Coding
- An Extended Kalman Filter for RTF Estimation in Dual-Microphone Smartphones
- Model Based Estimation of STP Parameters for Binaural Speech Enhancement
- Piano Legato-Pedal Onset Detection Based on a Sympathetic Resonance Measure
- Randomly Sketched Sparse Subspace Clustering for Acoustic Scene Clustering
- Anomalous Sound Event Detection Based on WaveNet
- Performance Analysis of the Covariance-Whitening and the Covariance-Subtraction Methods for Estimating the Relative Transfer Function
- Operational Rate-Constrained Beamforming in Binaural Hearing Aids
- Modulation-Domain Parametric Multichannel Kalman Filtering for Speech Enhancement
- Generative Adversarial Network-Based Approach to Signal Reconstruction from Magnitude Spectrogram
- Bitrate and Tandem Detection for the AMR-WB Codec with Application to Network Testing
- Anomaly Detection Based on Feature Reconstruction from Subsampled Audio Signals
- Perceived Quality of Audio-Visual Stimuli Containing Streaming Audio Degradations
- Using Acoustic Parameters for Intelligibility Prediction of Reverberant Speech
- Detection of North Atlantic Right Whales with a Hybrid System of CNN and Dictionary Learning
- Localization in Elevation with Non-Individual Head-Related Transfer Functions: Comparing Predictions of Two Auditory Models
- GEVD Based Speech and Noise Correlation Matrix Estimation for Multichannel Wiener Filter Based Noise Reduction
- Exterior and Interior Sound Field Separation Using Convex Optimization: Comparison of Signal Models
IFS-L1: Information forensics and security
Room: Auditorium
Chair: Slava Voloshynovskiy (University of Geneva, Switzerland, unknown)
- 10:30 Multiple Cyber-Threats Containment via Kendall's Birth-Death-Immigration Model
- 10:50 Blind Channel Direction Separation Against Pilot Spoofing Attack in Massive MIMO System
- 11:10 ADS-B Signal Signature Extraction for Intrusion Detection in the Air Traffic Surveillance System
- 11:30 Estimating Secret Parameters of a Random Number Generator from Time Series by Auto-synchronization
- 11:50 Authentication of Galileo GNSS Signal by Superimposed Signature with Artificial Noise
- 12:10 Privacy-Preserving Identification via Layered Sparse Code Design: Distributed Servers and Multiple Access Authorization
ML-L3: Machine Learning
Room Pininfarina
Chair: Constantine Kotropoulos (Aristotle University of Thessaloniki, Greece)
- 10:30 Robust Expectation Propagation in Factor Graphs Involving Both Continuous and Binary Variables
- 10:50 Prediction Methods for Time Evolving Dyadic Processes
- 11:10 Accelerated Stochastic Multiplicative Update with Gradient Averaging for Nonnegative Matrix Factorization
- 11:30 Decision Level Fusion: An Event Driven Approach
- 11:50 Entry-wise Matrix Completion from Noisy Entries
- 12:10 A Novel Formulation of Independence Detection Based on the Sample Characteristic Function
SPBD-L1: Signal Processing for Big Data
Room A
Chair: Danilo Mandic (Imperial College, London, United Kingdom (Great Britain))
- 10:30 High-Order CPD Estimation with Dimensionality Reduction Using A Tensor Train Model
- 11:00 Subspace-Orbit Randomized-Based Decomposition for Low-rank Matrix Approximations
- 11:30 Feature Fusion via Tensor Network Summation
- 12:00 A Novel Online Generalized Possibilistic Clustering Algorithm for Big Data Processing
SS-L14: Special Session on Advances in GW Signal Processing and Data Analysis
Room G
Chair: Paolo Addesso (University of Salerno, Italy)
Organizers: Paolo Addesso, Innocenzo M. Pinto, and Maria Principe
- 10:30 Techniques for Gravitational-Wave Detection of Compact Binary Coalescence
- 10:50 GPU-Optimised Low-Latency Online Search for Gravitational Waves from Binary Coalescences
- 11:10 Detection and Estimation of Unmodeled Chirps
- 11:30 Wavelet-based Classification of Transient Signals for Gravitational Wave Detectors
- 11:50 The Band-Sampled-Data Collection for the Search of Continuous Gravitational Wave Signals
- 12:10 Non-parametric Characterization of Gravitational-Wave Polarizations
SS-L15: Special Session on Signal Processing Advances for Radio Astronomy
Room Q
Chairs: Karim Abed-Meraim (Polytech'Orléans & University of Sharjah, UAE, France), Stefan J. Wijnholds (ASTRON, The Netherlands)
Organizers: Stefan J. Wijnholds and Karim Abed-Meraim
- 10:30 Beam Shape Calibration for Multi-Beam Radio Astronomical Phased Arrays
- 10:50 Uncertainty Quantification in Imaging When Convex Optimization Meets Bayesian Analysis
- 11:10 Bayesian Calibration Using Different Prior Distributions: An Iterative Maximum a Posteriori Approach for Radio Interferometers
- 11:30 Facet-Based Regularization for Scalable Radio-Interferometric Imaging
- 11:50 Radio Imaging with Information Field Theory
- 12:10 Efficient Calibration of Radio Interferometers Using Block LDU Decomposition
Friday, September 7 14:00 - 18:30
W1: WhoLoDance: Body motion analysis with applications to dance education and beyond
Room A
W2: STATOS thematic workshop 2018: Graph Signal Processing
Room G