The workshop will examine current opportunities and challenges for the signal and image sciences community. The forum is intended to enable a productive exchange of ideas on state-of-the-art technologies and recent developments. With the goal of sparking interesting discussions, we encourage submission of intermediate results from ongoing projects or recent conference papers. To enhance technical exchange between participants, we are also encouraging posters for this Workshop. The event is open to all engineers, scientists, and students with an interest in the signal and image sciences and is free of charge.
Among the broad range of technical topics slated are computed-tomography reconstruction and analysis, signal- and image processing and control in support of the National Ignition Facility, radiation detection and isotope ID, electromagnetic sensing methods, adaptive optics, novel sensing modalities (gravity gradiometry, biomedical and geophysical sensing), computer vision and video analytics, and machine learning and pattern analysis.
Mitubishi Electric Research Laboratories
Recent advances in inverse problems, including sparse signal recovery and non-convex optimization have shifted the design paradigm for sensing systems. Computational methods have become an integral part of the design toolbox, enabling the use of algorithms to address some of the hardware challenges in designing such systems. One of the most promising applications of this paradigm shift has been in array imaging systems, such as ultrasonic, radar and optical (LIDAR). The impact is also timely, as array processing is becoming increasingly important in a variety of applications, including robotics, autonomous driving, medical imaging, and virtual reality, among others. This has led to continuous improvements in sensing hardware, but also to increasing demand for theory and methods to inform the system design and improve the processing.
This talk will present a general inverse problem framework for array processing systems, which allows us to describe both the acquisition hardware and the scene being acquired. Under this framework we can exploit prior knowledge on the scene, the system, and the nature of a variety of errors that might occur, allowing for significant improvements in the reconstruction accuracy. Furthermore, we can consider the design of the system itself in the context of the inverse problem, leading to designs that are more efficient, more accurate, or less expensive, depending on the application. We will explore applications of this model such as LIDAR and depth sensing, radar and distributed radar, and ultrasonic sensing. In the context of these applications, we will describe how different models can lead to improved specifications in radar and ultrasonic systems, robustness to position and timing errors in distributed array systems, and cost reduction and new capabilities in LIDAR systems.
Petros T. Boufounos is a Distinguished Research Scientist and the Computational Sensing Team Leader at Mitsubishi Electric Research Laboratories (MERL). Dr. Boufounos completed his undergraduate and graduate studies at MIT. He received the S.B. degree in Economics in 2000, the S.B. and M.Eng. degrees in Electrical Engineering and Computer Science (EECS) in 2002, and the Sc.D. degree in EECS in 2006. Between September 2006 and December 2008, he was a postdoctoral associate with the Digital Signal Processing Group at Rice University. Dr. Boufounos joined MERL in January 2009, where he has been heading the Computational Sensing Team since 2016.
Dr. Boufounos' immediate research focus includes signal acquisition and processing, computational sensing, inverse problems, quantization, and data representations. He is also interested in how signal acquisition interacts with other fields that use sensing extensively, such as machine learning, robotics, and dynamical system theory. Dr. Boufounos has served as an Area Editor and a Senior Area Editor for the IEEE signal processing letters. He has been a part of the SigPort editorial board and is currently a member of the IEEE Signal Processing Society Theory and Methods technical committee and and an Associate Editor at IEEE Transactions on Computational Imaging. He was also named IEEE SPS Distinguished Lecturer for 2019-2020.
Spoofing attacks have been acknowledged as a serious threat to automatic speaker verification (ASV) systems. In this paper, we are specifically concerned with replay attack scenarios. As a countermeasure to the problem, we propose a front-end based on the blind estimation of the channel response magnitude and as a back-end a residual neural network. Our hypothesis is that the magnitude response of the channel, obtained by subtracting the log-magnitude spectrum of the observed signal from the prediction of the log-magnitude spectrum average of the observed signalβs clean counterpart, will capture the nuances of room ambiences, recordings and playback devices. The performance of these features is investigated on a benchmark back-end, based on a Gaussian mixture model and on a deep neural network classifier. Our experiments are performed on the 2017 and 2019 Automatic Speaker Verification Spoofing and Countermeasures Challenge (ASVspoof) datasets. The benchmark systems are the same as used in the challenges and are based on constant-Q cepstral coefficients (CQCC) and linear-frequency cepstral coefficients (LFCC) features. Experimental results on the 2017 dataset show that the proposed method outperforms the two benchmarks, providing equal-error rates (EER) as low as 7.57% and 11.64%, respectively, for the development and evaluation sets. On the ASVspoof 2019 dataset, in turn, the proposed method outperformed the benchmark using a residual neural network as back-end by yielding tandem detection cost function (t-DCF) and EER as low as 0.1086 and 4.26% on the evaluation set. Lastly, an instrumental (objective) quality assessment is performed on the two datasets and the impact of quality variability on spoofing detection accuracy is discussed.
Dr. Anderson finished his PhD at INRS, in September 2021. His main research lies on speech quality assessment, voice biometrics and speech emotion recognition. During his PhD, Anderson proposed two new speech quality measures based on i-vectors. He has also worked on the robustness of speech-based technologies "in-the-wild," being his main concern the performance of such solutions in real-world scenarios, where factors such as varying accent, emotions, vocal effort, background noise and reverberation can be detrimental. After finishing his PhD, Anderson joined Huawei Canada as a Machine Learning Scientist and is currently working on low-latency spoken language understanding.
Scientists and engineers are increasingly applying deep neural networks (DNNs) to modelling and design of complex systems. While the flexibility of DNNs makes them an attractive tool, it also makes their solutions difficult to interpret and their predictive capability difficult to quantify. In contrast, scientific models directly expose the equations governing a process but their applicability is restricted in the presence of unknown effects or when the data are high-dimensional. The emerging paradigm of physics-guided artificial intelligence asks: How can we combine the flexibility of DNNs with the interpretability of scientific models to learn relationships from data consistent with known scientific theories? In this talk, I will discuss my work on incorporating prior knowledge of problem structure (e.g., physics-based constraints) into neural network design. Specifically, I will demonstrate how prior knowledge of task symmetries can be leveraged for improved learning outcomes in convolutional neural network based classification; and how embedding priors from dynamical systems theory can lead to physically plausible neural network based video prediction.
Dr. Christine Allen-Blanchette is a postdoctoral researcher in the Department of Mechanical and Aerospace Engineering at Princeton University where she is pursuing research at the intersection of deep learning, geometry, and dynamical systems. Christine completed her PhD in Computer Science and MSE in Robotics at the University of Pennsylvania, and her BS degrees in Mechanical Engineering and Computer Engineering at San Jose State University. Among her awards are the Princeton Presidential Postdoctoral Fellowship, NSF Integrative Graduate Education and Research Training award, and GEM Fellowship sponsored by the Adobe Foundation.
SESSION β NIF Chair: Theresa Ann Green
|8:50β9:10am||Liliana Wang||Adaptive Optics optimization method with Hadamard basis for soil imaging|
|9:10β9:30am||Brad Funsten||Characterization of hCMOS Sensors for Flash X-ray Environments|
|9:30β9:50am||Erik Davies||Toward a 3D Velocity Interferometer Testbed: Concept and Algorithm Exploration|
SESSION β Computer Vision Chair: Alan David Kaplan
|9:50β10:10am||Rushil Anirudh||Generative Patch Priors for Compressive Image Recovery|
|10:10β10:30am||Aditya Mohan||AutoAtlas: Neural Network for 3D Unsupervised Partitioning and Representation Learning|
|10:30β10:50am||Amar Saini||Computational Visual Perception with Motion Boundary Sense and Conditional Attentive Latent Inference|
SESSION β Machine Learning Chairs: Hiranmayi Ranganathan, Ruben Glatt
|10:50β11:10am||Bhavya Kaikhura||Making AI Foolproof for Mission-Critical Applications|
|11:10β11:30am||Mikel Landajuela||Deep Symbolic Optimization: A framework for symbolic optimization using Deep Learning|
|11:30β11:50am||Brian Giera||Autonomous Multimodal Manufacturing Optimization|
|11:50amβ12:20pm||Dr. Anderson Avila||On the use of blind channel response estimation and a residual neural network to detect physical access attacks to speaker verification systems|
|12:20β12:50pm||Dr. Christine Allen-Blanchette||Leveraging Dataset Structure for Neural Network Prediction|
|8:50β9:50am||Dr. Petros Boufounos||The Computational-Sensing Revolution in Array Processing|
SESSION β Oscillations, Vibrations, and Stuff Chair: Sean K. Lehman
|9:50β10:10am||Brian Worthmann||Modeling Buried Object Brightness and Visibility for Ground Penetrating Radar|
|10:10β10:30am||Yaniv Rosen||The LLNL Quantum Design and Integration Testbed|
|10:30β10:50am||Saptarshi Mukherjee||Electrical Impedance Tomography Based on Adjoint Field-Sensitivity for Damage Detection in Highly Conductive Additively Manufactured Metal Mesh Structures|
|10:50β11:10am||Sean Lehman||A Broadband Multistatic Radar for Trajectory Indentification of Multiple Small Caliber Targets|
SESSION β NDE Chair: Kyle M. Champley
|11:10β11:30am||Rosa Morales||Real-time Laster Ultrasonic Monitoring of Laser-inducing Heating and Melting Processes|
|11:30β11:50am||Haichao Miao||Virtual Inspections: Visualization and Analysis of Complex Parts in Virtual Reality|
|11:50amβ12:10pm||Venkatesh Sridhar||Fast model-based algorithms for imaging through atmospheric turbulence|
|12:11β12:14pm||David Erskine||Toward a 3D Velocity Interferometer Testbed: Early Results|
|12:14β12:17pm||Sam Ade Jacobs||Learning to learn at HPC scale|
|12:17β12:20pm||Aditya Mohan||Constrained Non-Linear Phase Retrieval for Single Distance X-ray Phase Contrast Tomography|
|1:00β3:00pm||Virtual Poster Session [ ENTER HERE ]|
Abstract deadline: Wednesday, July 7th
Speaker confirmation: TBD
Proposal submission: Please email title and short abstract to Kyle Champley (firstname.lastname@example.org). Abstracts are used to organize sessions and will not be published.