2019 Western New York Image and Signal Processing
Workshop
The Western New York Image and
Signal Processing Workshop (WNYISPW) is a venue for
promoting image and signal processing research in our area
and for facilitating interaction between academic
researchers, industry researchers, and students. The
workshop comprises both oral and poster presentations.
The workshop, building off of 21 successful years of the Western New York Image Processing Workshop (WNYIPW), is sponsored by the Rochester chapter of the IEEE Signal Processing Society with technical cooperation from the Rochester chapter of the Society for Imaging Science and Technology.
The workshop will be held on Friday, October 04, 2019, in Louise Slaughter Hall (Building SLA/078) at Rochester Institute of Technology in Rochester, NY.
Topics
Topics include, but are not limited to:
Important Dates
Dr. David Doermann, University at Buffalo
"Media Manipulation and its Threat on Democracy"
Abstract:
The computer vision community has created a technology which unfortunately is getting more bad press then it is good. In 2014, the first GANS paper was able to automatically generate very low resolutions of faces of people which never existed, from a random latent distribution. Although the technology was impressive because it was automated, it was nowhere near as good as what could be done with the simple photo editor. In the same year DARPA started the media forensics program to combat the proliferation of edited images and video that was benign generated by our adversaries. Although DARPA envisioned the development automated technologies, no one thought they would evolve so fast. Five years later the technology has progressed to the point where even a novice can modify full videos, i.e. DeepFakes, and generate new content of people and scenes that never existed, overnight using commodity hardware. Recently the US government has become increasingly concerned about the real dangers of the use of “DeepFakes” technologies from both a national security and a misinformation point of view. To this end, it is important for academia, industry and the government to come together to apply technologies, develop policies that put pressure on service providers, and educate the public before we get to the point where “seeing is believing” is a thing of the past. In this talk I will cover some of the primary efforts in applying counter manipulation detection technology, the challenges we face with current policy in the United States. While technological solutions are still a number of years away, we need a comprehensive approach to deal with this problem.
Bio:
Dr. David Doermann is a Professor of Empire Innovation and the Director of the Artificial Intelligence Institute the University at Buffalo (UB). Prior to coming to UB he was a Program Manager with the Information Innovation Office at the Defense Advanced Research Projects Agency (DARPA) where he developed, selected and oversaw research and transition funding in the areas of computer vision, human language technologies and voice analytics. From 1993 to 2018, David was a member of the research faculty at the University of Maryland, College Park. In his role in the Institute for Advanced Computer Studies, he served as Director of the Laboratory for Language and Media Processing, and as an adjunct member of the graduate faculty for the Department of Computer Science and the Department of Electrical and Computer Engineering. He and his group of researchers focus on many innovative topics related to analysis and processing of document images and video including triage, visual indexing and retrieval, enhancement and recognition of both textual and structural components of visual media. David has over 250 publications in conferences and journals, is a fellow of the IEEE and IAPR, has numerous awards including an honorary doctorate from the University of Oulu, Finland and is a founding Editor-in-Chief of the International Journal on Document Analysis and Recognition.
Dr. Andrew Gallagher, Google
"Embracing Uncertainty: Knowing When We Don't Know"
Abstract:
In computer vision, instance embeddings are used to place an input sample (e.g., a face image) that has a large number of dimensions, into a more compact representation called an embedding. For example, one can represent an image as 128-dimensional point embedding, and the similarity between two images is related to the distance between their respective embeddings. But how do we know if an embedding is *good* or *bad*? Once a point embedding is computed, downstream users have no idea whether it is a good embedding (more likely than average to match with other faces with the same ID), or a bad embedding (more likely than average to match with strangers). Modeling uncertainty in embeddings is directed at quantifying the level of trust that we can place in a particular image's embedding. For example, if an image is highly blurred or heavily occluded, we would expect it to be less recognizable and hence the certainty should be lower. This talk will describe some uses of embeddings, and steps that we can take to address uncertainty.
Bio:
I joined Google in 2014. Previously, I was a Visiting Research Scientist at Cornell University's School of Electrical and Computer Engineering, beginning in June 2012. I earned the Ph.D. degree in electrical and computer engineering from Carnegie Mellon University in 2009, advised by Prof. Tsuhan Chen. I received an M.S. degree from Rochester Institute of Technology, and the B.S. degree from Geneva College, both in electrical engineering. I worked for the Eastman Kodak Company for over a decade during the fascinating transition from chemical to digital imaging, initially developing image enhancement algorithms for digital photofinishing. These algorithms were shipped under the trade name "Kodak Perfect Touch" in photo printing mini-labs, and millions of digital cameras, and enhanced many billions of images. I enjoy working on tough and interesting problems.
Presentation slides: here.
Invited Speakers
Mujdat Cetin -- Dept. of ECE, University of Rochester
"Compressed Sensing and Machine Learning for Radar Imaging"
Abstract:
In this talk we first present an overview of our past work that lies at the intersection of two domains: sparse signal representation and computational radar imaging, in particular synthetic aperture radar (SAR) image formation. We present a probabilistic perspective for SAR imaging, which leads to optimization-based image formation algorithms. Within this framework, we describe how sparsity-driven priors or regularization constraints emerge as useful notions for solution of ill-posed radar imaging problems, leading to compressed sensing methods. We present examples demonstrating the effectiveness of this perspective in a variety of SAR imaging scenarios. Next, we turn to more recent work that brings in machine learning into the process of SAR image formation. As an initial attempt, we describe dictionary learning methods aimed to tune sparsity-driven priors to a particular context. Then, we present a framework in which we incorporate convolutional neural network (CNN) based prior models into SAR image formation. In particular, we propose a plug-and-play (PnP) priors based approach that allows joint use of physics-based forward models and state-of-the-art prior models. We demonstrate preliminary results demonstrating the potential of this machine learning based approach for the reconstruction of synthetic and real SAR scenes.
Bio:
Mujdat Cetin is an Associate Professor of Electrical and Computer Engineering and the Interim Director of the Goergen Institute for Data Science at the University of Rochester. From 2005-2017, he was a faculty member at Sabanci University, Istanbul, Turkey. From 2001 to 2005, he was with the Laboratory for Information and Decision Systems, MIT. Prof. Cetin has held visiting faculty positions at MIT, Northeastern University, and Boston University. Prof. Cetin’s research interests are within the broad area of data, signal, and imaging sciences, with cross-disciplinary links to several other areas in electrical engineering, computer science, and neuroscience. His research group has made advances in three key areas: computational sensing and imaging as applied to radar and biomedical imaging; probabilistic methods for image and video analysis as applied to biomedical image analysis, microscopic neuroimaging, and computer vision; and signal processing and machine learning for brain-computer/machine interfaces. Prof. Cetin is a Fellow of the IEEE and has received several awards including the IEEE Signal Processing Society Best Paper Award; the EURASIP/Elsevier Signal Processing Best Paper Award; the IET Radar, Sonar and Navigation Premium Award; and the Turkish Academy of Sciences Distinguished Young Scientist Award. Prof. Cetin is currently the Chair of the IEEE Computational Imaging Technical Committee. He is also a Senior Area Editor for the IEEE Transactions on Computational Imaging and for the IEEE Transactions on Image Processing. Prof. Cetin was the Technical Program Co-chair for the IEEE Image, Video, and Multidimensional Signal Processing (IVMSP) Workshop in 2016; for the International Conference on Information Fusion in 2016 and 2013; for the International Conference on Pattern Recognition (ICPR) in 2010; and for the IEEE Turkish Conference on Signal Processing, Communications, and their Applications in 2006. He was one of the keynote speakers for the 2015 International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar, and Remote Sensing.
Presentation slides: here.
Gonzalo Mateos -- Dept. of ECE, University of Rochester
"Graph Signal Processing: Foundational Advances for Learning from Network Data"
Abstract:
Coping with the challenges found at the intersection of Network Science and Big Data necessitates fundamental breakthroughs in modeling, identification, and controllability of distributed network processes – often conceptualized as signals defined on graphs. For instance, graph-supported signals can model vehicle trajectories over road networks; economic activity observed over a network of production flows between industrial sectors; infectious states of individuals susceptible to an epidemic disease spreading on a social network; brain activity signals supported on brain functional connectivity networks; and media cascades that diffuse on online social networks, to name a few. There is an evident mismatch between our scientific understanding of signals defined over regular domains (time or space) and graph-valued signals. Knowledge about time series was developed over the course of decades and boosted by real needs in areas such as communications, speech, or control. On the contrary, the prevalence of network-related information processing problems and the access to quality network data are recent events. Under the natural assumption that the signal properties are related to the topology of the graph where they are supported, the emerging field of graph signal processing (GSP) aims at developing data science algorithms that fruitfully leverage this relational structure, and can make inferences about these relationships when they are only partially observed. In this talk I will start by presenting the fundamentals of GSP and motivating the study of graph signals. Then I will leverage these ideas to offer a fresh look at the problems of graph learning, source localization on graphs, and orthonormal (Fourier like) signal representations. Throughout, we illustrate the developed methods and results on various application domains including urban mobility, the economy, network neuroscience, and environmental monitoring.
Bio:
Gonzalo Mateos earned the B.Sc. degree from Universidad de la Republica, Uruguay, in 2005, and the M.Sc. and Ph.D. degrees from the University of Minnesota, Twin Cities, in 2009 and 2011, all in electrical engineering. He joined the University of Rochester, Rochester, NY, in 2014, where he is currently an Assistant Professor with the Department of Electrical and Computer Engineering, as well as a member of the Goergen Institute for Data Science. During the 2013 academic year, he was a visiting scholar with the Computer Science Department at Carnegie Mellon University. From 2004 to 2006, he worked as a Systems Engineer at Asea Brown Boveri (ABB), Uruguay. His research interests lie in the areas of statistical learning from Big Data, network science, decentralized optimization, and graph signal processing, with applications in dynamic network health monitoring, social, power grid, and Big Data analytics. He currently serves as Associate Editor for the IEEE Transactions on Signal Processing, the IEEE Transactions on Signal and Information Processing over Networks, and is a member of the IEEE SigPort Editorial Board. Dr. Mateos received the NSF CAREER Award in 2018, the 2017 IEEE Signal Processing Society Young Author Best Paper Award (as senior co-author), and the Best Paper Awards at SPAWC 2012, SSP 2016, as well as ICASSP 2018 and 2019. His doctoral work has been recognized with the 2013 University of Minnesota's Best Dissertation Award (Honorable Mention) across all Physical Sciences and Engineering areas.
Presentation slides: here.
Rob Phipps, VisualDx
"Machine Learning for Physician Assistance in Dermatology"
Abstract:
How do you start with a couple of guys who have watched the MIT 6.034 lectures ( the homework? Ha!), a couple of GPUs, and a pile of images collected over the years for different purposes and end up with a system that assists clinicians in diagnosing skin issues? Well, there are a lot of pitfalls along the way and a few lessons learned that span dermatology, image science, data science, computer science, and software engineering. This isn’t a theoretical talk, this the experience of learning just enough to get things to work well enough to get into a product while still doing your day job. Not the textbook way of doing things, but the way this technology is playing out in lots of small companies today.
Bio:
Rob Phipps is Vice President for Engineering at VisualDx, a world-leading healthcare company in the field of healthcare informatics. At VisualDx, Rob is on the executive leadership team, responsible for product and software development. Earlier in his career, he worked in Kodak's Research Labs, helping define JAI and ColorSense along with other imaging and color science software frameworks. Later, as a consultant for almost 20 years, he did imaging software work for Sun MicroSystems, JPL, Lockheed Martin, and others. Rob is a graduate of the Massachusetts Institute of Technology).
Junsong Yuan, University of Buffalo
"Hand Sensing for Augmented Interaction"
Abstract:
Hands are crucial for humans to interact with physical world. In social interactions, we also rely on our hand gestures to communicate with each other. In this talk, I will discuss real-time human hand sensing using different types of cameras, such as optical camera, depth camera, and event camera. I will also discuss how to leverage synthetic data to address the high-dimensional regression problem of articulated hand pose estimation in real time. The resulting systems can facilitate intelligent interactions in virtual and real environments using bare hands. Moreover, with the understanding of hand gestures, robots can better sense the intentions of humans, and communicate with them in a more natural way. Applications in virtual/augmented reality, tele-operation, human-robot interaction will also be discussed.
Bio:
Junsong Yuan is an Associate Professor and Director of Visual Computing Lab at Department of Computer Science and Engineering, State University of New York at Buffalo, USA. Before that he was an Associate Professor at Nanyang Technological University (NTU), Singapore. He received his PhD from Northwestern University and M.Eng. from National University of Singapore. He is currently Senior Area Editor of Journal of Visual Communications and Image Representation (JVCI), Associate Editor of IEEE Trans. on Image Processing (T-IP), IEEE Trans. on Circuits and Systems for Video Technology (T-CSVT), and The Visual Computer (TVC). He also served as Area Chair for CVPR, ICIP, ICPR, ACCV, ACM MM, WACV etc. He received 2016 Best Paper Award from IEEE Trans. on Multimedia, Nanyang Assistant Professorship from NTU, and Outstanding EECS Ph.D. Thesis award from Northwestern University. He is a Fellow of International Association of Pattern Recognition (IAPR).
Presentation slides: here.
Tutorials
Jianghao Wang, Data Scienist, Data Scientist
Div Tiwari, Customer Success Engineer
Alyssa Silverman, Field Engineer and Account Manager
"Demystifying deep learning: A practical approach in MATLAB"
Abstract:
Are you new to deep learning and want to learn how to use it in your work? Deep learning can achieve state-of-the-art accuracy in many humanlike tasks such as naming objects in a scene or recognizing optimal paths in an environment.
The main tasks are to assemble large data sets, create a neural network, to train, visualize, and evaluate different models, using specialized hardware - often requiring unique programming knowledge. These tasks are frequently even more challenging because of the complex theory behind them.
In this seminar, we’ll demonstrate new MATLAB features that simplify these tasks and eliminate the low-level programming. In doing so, we’ll decipher practical knowledge of the domain of deep learning. We’ll build and train neural networks that recognize handwriting, classify food in a scene, classify signals, and figure out the drivable area in a city environment.
Along the way, you’ll see MATLAB features that make it easy to:
Notes: This is a seminar style presentation and there is no prior preparation from the attendees.
Sidney Pendelberry, RIT Research Computing
"Research Computing Tutorial"
Abstract:
This workshop is an introduction to using high-performance computing (HPC) when your compute needs exceed your workstations and lab computers. This overview focuses on the tools needed to get started in an HPC environment. Topics covered are: job submission, data storage and transfer, loading software, job arrays, package manager (spack), working with gpus, as well as examples running: C, R, Matlab, and Python.
Paper Submission
The Call for Papers can be found here.
Paper submission is now closed! Paper submission deadline 11:59pm EDT on September 13, 2019.
--> Prospective authors are invited to submit a 4-page paper + 5th page of references of previously unpublished paper for oral or poster presentation. here: https://cmt3.research.microsoft.com/WNYISPW2019/
Authors should use the same formatting/templates described in the ICIP 2019 Paper Kit. Note, this is a single blind submission and author names and affiliations should be included on submitted papers.
Conference content will be submitted for inclusion into IEEE Xplore as well as other Abstracting and Indexing (A&I) databases. Past WNYIPW and WNYISPW proceedings can be found here:
--> Authors who only want to be considered for a poster presentaiton have the option to submit an abstract in place of a full paper. This may be suitable for recent work that is not quite ready for publication or has already been published. (Note: Abstract-only submissions will not be searchable on IEEE Xplore)
Prospective authors are invited to submit an abstract here: https://cmt3.research.microsoft.com/WNYISPW2019/
Awards
To encourage student participation, a best student paper and best student poster
award will be given.
The workshop, building off of 21 successful years of the Western New York Image Processing Workshop (WNYIPW), is sponsored by the Rochester chapter of the IEEE Signal Processing Society with technical cooperation from the Rochester chapter of the Society for Imaging Science and Technology.
The workshop will be held on Friday, October 04, 2019, in Louise Slaughter Hall (Building SLA/078) at Rochester Institute of Technology in Rochester, NY.
Topics
Topics include, but are not limited to:- Formation, Processing, and/or Analysis of Signals, Images, or Video
- Computer Vision
- Information Retrieval
- Image and Color Science
- Applications of Image and Signal Processing, including:
- Medical Image and Signal Analysis
- Audio Processing and Analysis
- Remote Sensing
- Archival Imaging
- Printing
- Consumer Devices
- Security
- Surveillance
- Document Imaging
- Art Restoration and Analysis
- Astronomy
Important Dates
| Paper/poster submission opens: | August 13, 2019 |
| Paper submission closes: | September 13, 2019 |
| Poster submission closes: | September 20, 2019 |
| Notification of Acceptance: | September 18, 2019 |
| Early (online) registration deadline: | September 20, 2019 |
| Submission of camera-ready paper: | October 7, 2019 |
| Workshop: | October 04, 2019 |
Keynote Speakers
We are happy to announce our keynote speakers:"Media Manipulation and its Threat on Democracy"
Abstract:
The computer vision community has created a technology which unfortunately is getting more bad press then it is good. In 2014, the first GANS paper was able to automatically generate very low resolutions of faces of people which never existed, from a random latent distribution. Although the technology was impressive because it was automated, it was nowhere near as good as what could be done with the simple photo editor. In the same year DARPA started the media forensics program to combat the proliferation of edited images and video that was benign generated by our adversaries. Although DARPA envisioned the development automated technologies, no one thought they would evolve so fast. Five years later the technology has progressed to the point where even a novice can modify full videos, i.e. DeepFakes, and generate new content of people and scenes that never existed, overnight using commodity hardware. Recently the US government has become increasingly concerned about the real dangers of the use of “DeepFakes” technologies from both a national security and a misinformation point of view. To this end, it is important for academia, industry and the government to come together to apply technologies, develop policies that put pressure on service providers, and educate the public before we get to the point where “seeing is believing” is a thing of the past. In this talk I will cover some of the primary efforts in applying counter manipulation detection technology, the challenges we face with current policy in the United States. While technological solutions are still a number of years away, we need a comprehensive approach to deal with this problem.
Bio:
Dr. David Doermann is a Professor of Empire Innovation and the Director of the Artificial Intelligence Institute the University at Buffalo (UB). Prior to coming to UB he was a Program Manager with the Information Innovation Office at the Defense Advanced Research Projects Agency (DARPA) where he developed, selected and oversaw research and transition funding in the areas of computer vision, human language technologies and voice analytics. From 1993 to 2018, David was a member of the research faculty at the University of Maryland, College Park. In his role in the Institute for Advanced Computer Studies, he served as Director of the Laboratory for Language and Media Processing, and as an adjunct member of the graduate faculty for the Department of Computer Science and the Department of Electrical and Computer Engineering. He and his group of researchers focus on many innovative topics related to analysis and processing of document images and video including triage, visual indexing and retrieval, enhancement and recognition of both textual and structural components of visual media. David has over 250 publications in conferences and journals, is a fellow of the IEEE and IAPR, has numerous awards including an honorary doctorate from the University of Oulu, Finland and is a founding Editor-in-Chief of the International Journal on Document Analysis and Recognition.
"Embracing Uncertainty: Knowing When We Don't Know"
Abstract:
In computer vision, instance embeddings are used to place an input sample (e.g., a face image) that has a large number of dimensions, into a more compact representation called an embedding. For example, one can represent an image as 128-dimensional point embedding, and the similarity between two images is related to the distance between their respective embeddings. But how do we know if an embedding is *good* or *bad*? Once a point embedding is computed, downstream users have no idea whether it is a good embedding (more likely than average to match with other faces with the same ID), or a bad embedding (more likely than average to match with strangers). Modeling uncertainty in embeddings is directed at quantifying the level of trust that we can place in a particular image's embedding. For example, if an image is highly blurred or heavily occluded, we would expect it to be less recognizable and hence the certainty should be lower. This talk will describe some uses of embeddings, and steps that we can take to address uncertainty.
Bio:
I joined Google in 2014. Previously, I was a Visiting Research Scientist at Cornell University's School of Electrical and Computer Engineering, beginning in June 2012. I earned the Ph.D. degree in electrical and computer engineering from Carnegie Mellon University in 2009, advised by Prof. Tsuhan Chen. I received an M.S. degree from Rochester Institute of Technology, and the B.S. degree from Geneva College, both in electrical engineering. I worked for the Eastman Kodak Company for over a decade during the fascinating transition from chemical to digital imaging, initially developing image enhancement algorithms for digital photofinishing. These algorithms were shipped under the trade name "Kodak Perfect Touch" in photo printing mini-labs, and millions of digital cameras, and enhanced many billions of images. I enjoy working on tough and interesting problems.
Presentation slides: here.
Invited Speakers
"Compressed Sensing and Machine Learning for Radar Imaging"
Abstract:
In this talk we first present an overview of our past work that lies at the intersection of two domains: sparse signal representation and computational radar imaging, in particular synthetic aperture radar (SAR) image formation. We present a probabilistic perspective for SAR imaging, which leads to optimization-based image formation algorithms. Within this framework, we describe how sparsity-driven priors or regularization constraints emerge as useful notions for solution of ill-posed radar imaging problems, leading to compressed sensing methods. We present examples demonstrating the effectiveness of this perspective in a variety of SAR imaging scenarios. Next, we turn to more recent work that brings in machine learning into the process of SAR image formation. As an initial attempt, we describe dictionary learning methods aimed to tune sparsity-driven priors to a particular context. Then, we present a framework in which we incorporate convolutional neural network (CNN) based prior models into SAR image formation. In particular, we propose a plug-and-play (PnP) priors based approach that allows joint use of physics-based forward models and state-of-the-art prior models. We demonstrate preliminary results demonstrating the potential of this machine learning based approach for the reconstruction of synthetic and real SAR scenes.
Bio:
Mujdat Cetin is an Associate Professor of Electrical and Computer Engineering and the Interim Director of the Goergen Institute for Data Science at the University of Rochester. From 2005-2017, he was a faculty member at Sabanci University, Istanbul, Turkey. From 2001 to 2005, he was with the Laboratory for Information and Decision Systems, MIT. Prof. Cetin has held visiting faculty positions at MIT, Northeastern University, and Boston University. Prof. Cetin’s research interests are within the broad area of data, signal, and imaging sciences, with cross-disciplinary links to several other areas in electrical engineering, computer science, and neuroscience. His research group has made advances in three key areas: computational sensing and imaging as applied to radar and biomedical imaging; probabilistic methods for image and video analysis as applied to biomedical image analysis, microscopic neuroimaging, and computer vision; and signal processing and machine learning for brain-computer/machine interfaces. Prof. Cetin is a Fellow of the IEEE and has received several awards including the IEEE Signal Processing Society Best Paper Award; the EURASIP/Elsevier Signal Processing Best Paper Award; the IET Radar, Sonar and Navigation Premium Award; and the Turkish Academy of Sciences Distinguished Young Scientist Award. Prof. Cetin is currently the Chair of the IEEE Computational Imaging Technical Committee. He is also a Senior Area Editor for the IEEE Transactions on Computational Imaging and for the IEEE Transactions on Image Processing. Prof. Cetin was the Technical Program Co-chair for the IEEE Image, Video, and Multidimensional Signal Processing (IVMSP) Workshop in 2016; for the International Conference on Information Fusion in 2016 and 2013; for the International Conference on Pattern Recognition (ICPR) in 2010; and for the IEEE Turkish Conference on Signal Processing, Communications, and their Applications in 2006. He was one of the keynote speakers for the 2015 International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar, and Remote Sensing.
Presentation slides: here.
"Graph Signal Processing: Foundational Advances for Learning from Network Data"
Abstract:
Coping with the challenges found at the intersection of Network Science and Big Data necessitates fundamental breakthroughs in modeling, identification, and controllability of distributed network processes – often conceptualized as signals defined on graphs. For instance, graph-supported signals can model vehicle trajectories over road networks; economic activity observed over a network of production flows between industrial sectors; infectious states of individuals susceptible to an epidemic disease spreading on a social network; brain activity signals supported on brain functional connectivity networks; and media cascades that diffuse on online social networks, to name a few. There is an evident mismatch between our scientific understanding of signals defined over regular domains (time or space) and graph-valued signals. Knowledge about time series was developed over the course of decades and boosted by real needs in areas such as communications, speech, or control. On the contrary, the prevalence of network-related information processing problems and the access to quality network data are recent events. Under the natural assumption that the signal properties are related to the topology of the graph where they are supported, the emerging field of graph signal processing (GSP) aims at developing data science algorithms that fruitfully leverage this relational structure, and can make inferences about these relationships when they are only partially observed. In this talk I will start by presenting the fundamentals of GSP and motivating the study of graph signals. Then I will leverage these ideas to offer a fresh look at the problems of graph learning, source localization on graphs, and orthonormal (Fourier like) signal representations. Throughout, we illustrate the developed methods and results on various application domains including urban mobility, the economy, network neuroscience, and environmental monitoring.
Bio:
Gonzalo Mateos earned the B.Sc. degree from Universidad de la Republica, Uruguay, in 2005, and the M.Sc. and Ph.D. degrees from the University of Minnesota, Twin Cities, in 2009 and 2011, all in electrical engineering. He joined the University of Rochester, Rochester, NY, in 2014, where he is currently an Assistant Professor with the Department of Electrical and Computer Engineering, as well as a member of the Goergen Institute for Data Science. During the 2013 academic year, he was a visiting scholar with the Computer Science Department at Carnegie Mellon University. From 2004 to 2006, he worked as a Systems Engineer at Asea Brown Boveri (ABB), Uruguay. His research interests lie in the areas of statistical learning from Big Data, network science, decentralized optimization, and graph signal processing, with applications in dynamic network health monitoring, social, power grid, and Big Data analytics. He currently serves as Associate Editor for the IEEE Transactions on Signal Processing, the IEEE Transactions on Signal and Information Processing over Networks, and is a member of the IEEE SigPort Editorial Board. Dr. Mateos received the NSF CAREER Award in 2018, the 2017 IEEE Signal Processing Society Young Author Best Paper Award (as senior co-author), and the Best Paper Awards at SPAWC 2012, SSP 2016, as well as ICASSP 2018 and 2019. His doctoral work has been recognized with the 2013 University of Minnesota's Best Dissertation Award (Honorable Mention) across all Physical Sciences and Engineering areas.
Presentation slides: here.
"Machine Learning for Physician Assistance in Dermatology"
Abstract:
How do you start with a couple of guys who have watched the MIT 6.034 lectures ( the homework? Ha!), a couple of GPUs, and a pile of images collected over the years for different purposes and end up with a system that assists clinicians in diagnosing skin issues? Well, there are a lot of pitfalls along the way and a few lessons learned that span dermatology, image science, data science, computer science, and software engineering. This isn’t a theoretical talk, this the experience of learning just enough to get things to work well enough to get into a product while still doing your day job. Not the textbook way of doing things, but the way this technology is playing out in lots of small companies today.
Bio:
Rob Phipps is Vice President for Engineering at VisualDx, a world-leading healthcare company in the field of healthcare informatics. At VisualDx, Rob is on the executive leadership team, responsible for product and software development. Earlier in his career, he worked in Kodak's Research Labs, helping define JAI and ColorSense along with other imaging and color science software frameworks. Later, as a consultant for almost 20 years, he did imaging software work for Sun MicroSystems, JPL, Lockheed Martin, and others. Rob is a graduate of the Massachusetts Institute of Technology).
"Hand Sensing for Augmented Interaction"
Abstract:
Hands are crucial for humans to interact with physical world. In social interactions, we also rely on our hand gestures to communicate with each other. In this talk, I will discuss real-time human hand sensing using different types of cameras, such as optical camera, depth camera, and event camera. I will also discuss how to leverage synthetic data to address the high-dimensional regression problem of articulated hand pose estimation in real time. The resulting systems can facilitate intelligent interactions in virtual and real environments using bare hands. Moreover, with the understanding of hand gestures, robots can better sense the intentions of humans, and communicate with them in a more natural way. Applications in virtual/augmented reality, tele-operation, human-robot interaction will also be discussed.
Bio:
Junsong Yuan is an Associate Professor and Director of Visual Computing Lab at Department of Computer Science and Engineering, State University of New York at Buffalo, USA. Before that he was an Associate Professor at Nanyang Technological University (NTU), Singapore. He received his PhD from Northwestern University and M.Eng. from National University of Singapore. He is currently Senior Area Editor of Journal of Visual Communications and Image Representation (JVCI), Associate Editor of IEEE Trans. on Image Processing (T-IP), IEEE Trans. on Circuits and Systems for Video Technology (T-CSVT), and The Visual Computer (TVC). He also served as Area Chair for CVPR, ICIP, ICPR, ACCV, ACM MM, WACV etc. He received 2016 Best Paper Award from IEEE Trans. on Multimedia, Nanyang Assistant Professorship from NTU, and Outstanding EECS Ph.D. Thesis award from Northwestern University. He is a Fellow of International Association of Pattern Recognition (IAPR).
Presentation slides: here.
Tutorials
Div Tiwari, Customer Success Engineer
Alyssa Silverman, Field Engineer and Account Manager
"Demystifying deep learning: A practical approach in MATLAB"
Abstract:
Are you new to deep learning and want to learn how to use it in your work? Deep learning can achieve state-of-the-art accuracy in many humanlike tasks such as naming objects in a scene or recognizing optimal paths in an environment.
The main tasks are to assemble large data sets, create a neural network, to train, visualize, and evaluate different models, using specialized hardware - often requiring unique programming knowledge. These tasks are frequently even more challenging because of the complex theory behind them.
In this seminar, we’ll demonstrate new MATLAB features that simplify these tasks and eliminate the low-level programming. In doing so, we’ll decipher practical knowledge of the domain of deep learning. We’ll build and train neural networks that recognize handwriting, classify food in a scene, classify signals, and figure out the drivable area in a city environment.
Along the way, you’ll see MATLAB features that make it easy to:
- Manage large sets of images
- Create, analyze, and visualize networks and gain insight into the black box nature of deep networks
- Build networks from scratch with a drag-and-drop interface
- Perform classification tasks on images and signals, and pixel-level semantic segmentation on images
- Import training data sets from networks such as GoogLeNet and ResNet
- Import models from TensorFlow Keras, Caffe, and the ONNX Model format
- Speed up network training with parallel computing on a cluster
- Automate manual effort required to label ground truth
- Automatically generate source code for embedded targets
Notes: This is a seminar style presentation and there is no prior preparation from the attendees.
"Research Computing Tutorial"
Abstract:
This workshop is an introduction to using high-performance computing (HPC) when your compute needs exceed your workstations and lab computers. This overview focuses on the tools needed to get started in an HPC environment. Topics covered are: job submission, data storage and transfer, loading software, job arrays, package manager (spack), working with gpus, as well as examples running: C, R, Matlab, and Python.
Paper Submission
The Call for Papers can be found here.
Paper submission is now closed! Paper submission deadline 11:59pm EDT on September 13, 2019.
--> Prospective authors are invited to submit a 4-page paper + 5th page of references of previously unpublished paper for oral or poster presentation. here: https://cmt3.research.microsoft.com/WNYISPW2019/
Authors should use the same formatting/templates described in the ICIP 2019 Paper Kit. Note, this is a single blind submission and author names and affiliations should be included on submitted papers.
Conference content will be submitted for inclusion into IEEE Xplore as well as other Abstracting and Indexing (A&I) databases. Past WNYIPW and WNYISPW proceedings can be found here:
Poster Submission
Poster submission deadline 11:59pm EDT on September 20, 2019.--> Authors who only want to be considered for a poster presentaiton have the option to submit an abstract in place of a full paper. This may be suitable for recent work that is not quite ready for publication or has already been published. (Note: Abstract-only submissions will not be searchable on IEEE Xplore)
Prospective authors are invited to submit an abstract here: https://cmt3.research.microsoft.com/WNYISPW2019/
Author Attendance
At least one author of each accepted paper or poster must register and attend the workshop to give an oral or poster presentation. Failure to present the paper will result in automatic withdrawal of the paper from being published in the proceedings.Distant Authors
We will do our best to give early notification to those authors who need to make travel plans. No invitations will be given for international visas. At least one author needs to be present to prevent paper withdrawal from proceedings.Awards
To encourage student participation, a best student paper and best student poster
award will be given. Registration
Registration is available online here. Onsite registration will be also available, with onsite registration fees payable by cash or check. Fees enable attendance to all sessions and include breakfast, lunch, and afternoon snack. Registration fees are:- General Registration: $60 (with online registration by
09/20), $70 (online after 09/20 or onsite)
- Student Registration: $40 (with online registration by 09/20), $50 (online after 09/20 or onsite)
- IEEE or IS&T Membership: $40 (with online registration by 09/20), $50 (online after 09/20 or onsite)
- IEEE or IS&T Student Membership: $30 (with online registration by 09/20), $40 (online after 09/20 or onsite)
Conference at a Glance
- 8:30-8:50am, Registration, Breakfast
- 8:50-9am, Welcome by Chair
- 9-9:45am, Andrew Gallagher
- 9:45-10am, coffee break
- 10-11am, Oral presentations (4 presentations)
- 11-11:30am, Junsong Yuan
- 11:30am-Noon, Mujdat Cetin
- 10am-Noon, Deep learning tutorial by Mathwork’s Jianghao Wang (parallel track- Room 2120)
- Noon-1:30pm, Lunch and posters
- 1:30-2:15pm, David Doermann
- 2:15-2:30pm, snack break
- 2:30-3:30pm, Oral presentations (4 presentations)
- 3:30-4pm, Gonzalo Mateos Buckstein
- 4-4:30pm, Rob Phipps
- 2:30-4:30pm, RIT Research Computing by Sidney Pendelberry (parallel track – Room 2120)
- 4:30-4:45pm, Bryan Blakeslee 2018 Best paper presentation winner
- 4:45-5pm, Awards and wrapup
- 8:30-8:50am, Registration, Breakfast
- 8:50-9am, Welcome by Chair
- 9-9:45am, Keynote: Andrew Gallagher
- 9:45-10am, coffee break
- 10-11am, AM Oral presentations:
- 10 am - "SYSTEM SIGNALS MONITORING AND PROCESSING FOR COLLUDED APPLICATION ATTACKS DETECTION IN ANDROID OS", Igor Khokhlov (Rochester Institute of Technology); Michael Perez (Rochester Institute of Technology); Leon Reznik (Rochester Institute of Technology)
- 10:15 am - "EVOLUTION OF GRAPH CLASSIFIERS", Miguel Dominguez (Rochester Institute of Technology); Rohan N Dhamdhere (Rochester Institute of Technology); Naga Daga Harish Kanamarlapudi (Rochester Institute of Technology); Sunand Raghupathi (Columbia University); Raymond Ptucha (Rochester Institute of Technology) --Best paper award!
- 10:30 am - "Reduced Complexity Tree-Search Detector for Hybrid Space-Time Codes", Griselda Gonzalez (Instituto Tecnologico de Sonora); Joaquin Cortez (Instituto Tecnologico de Sonora); Miguel Bazdresch (Rochester Institute of Technology)
- 10:45 am - "Affective Video Recommender System", Yashowardhan Soni (Rochester Institute of Technology); Cecilia Alm (Rochester Institute of Technology); Reynold Bailey (Rochester Institute of Technology)
- 11-11:30am, Invited: Junsong Yuan
- 11:30am-Noon, Invited: Mujdat Cetin
- 10am-Noon, Deep learning tutorial by Mathwork’s Jianghao Wang (parallel track- Room 2120)
- Noon-1:30pm, Lunch and posters
- "Complex Neural Networks for Radio Frequency Fingerprinting", James M Stankowicz (BAE Systems); Scott Kuzdeba (BAE Systems); Joseph Carmack (BAE Systems); Joshua Robinson (BAE Systems)
- "A CCA APPROACH FOR MULTIVIEW ANALYSIS TO DETECT RIGID GAS PERMEABLE LENS BASE CURVE", Sara Hashemi (University of Tehran), Hadi Veisi (University of Tehran), Ebrahim Jafarzadehpur (Iran University of Medical Sciences), Rouhollah Rahmani (Digikala company), Zeinabolhoda Heshmati (University of Tehran)
- "A Data-Driven Model to Identify Fatigue Level Based on the Motion Data from a Smartphone",Swapnali Karvekar (Rochester Institute of Technology); Masoud Abdollahi (Rochester Institute of Technology); Ehsan Rashedi (Rochester Institute of Technology)
- "A COMPARISON OF CORRELATION PREDICTORS FOR PRNU-BASED IMAGE MANIPULATION LOCALIZATION",Sujoy Chakraborty (Stockton University)
- "Evaluating Security Metrics for Website Fingerprinting Defenses",Nate J Mathews (Rochester Institute of Technology); Saidur Rahman (Rochester Institute of Technology); Matthew Wright (Rochester Institute of Technology)
- "Quantitative High-Frequency Ultrasound for Characterizing Collagen Microstructure in Tendon",Sarah E Wayson (Department of Biomedical Engineering, University of Rochester); María Helguera (Tecnológico Mario Molina); Denise C. Hocking (Department of Pharmacology and Physiology, University of Rochester); Diane Dalecki (Department of Biomedical Engineering, University of Rochester)
- "Change Detection in Satellite Imagery with Region Proposal Networks",Navya Nagananda (Rochester Institute of Technology); Andreas Savakis (Rochester Institute of Technology)
- "Cherry: Sound localization for the DHH (deaf and hard of hearing)",Hrishikesh H Karale (Complemar); Gary Behm (Rochester Institute of Technology)
- "Vid2Cartoon: Turning Videos into Episodes of the Simpsons with Neural Animation Style Transfer", Max Lipitz (Rochester Institute of Technology); Christopher Kanan (Rochester Institute of Technology)
- "Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysis",Tyler Hayes (Rochester Institute of Technology); Christopher Kanan (Rochester Institute of Technology)
- "Implementing an efferent signal into an auditory pathway model for tone-in-noise detection",Afagh Farhadi (University of Rochester); Laurel H. Carney (University of Rochester)
- "Comparing Modality performance for Deep Video Summarization", Brendan M Wells (Rochester Institute of Technology); Alexander Loui (Rochester Institute of Technology)
- "Hand localization for hand grasp analysis in natural tasks",Anjali K Jogeshwar (Rochester Institute of Technology); Jeff Pelz (Rochester Institute of Technology) --Best poster award!
- "Towards Robust Open-World Detection of Deepfakes",Saniat Sohrawardi (Rochester Institute of Technology); Akash Chintha (Rochester Institute of Technology); Bao Thai (Rochester Institute of Technology); Raymond Ptucha (Rochester Institute of Technology); Matthew Wright (Rochester Institute of Technology)
- "Answer Them All! Toward Universal Visual Question Answering Models",Robik S Shrestha (Rochester Institute of Technology)
- "Incremental Open Set Recognition: Exploring Novel Input Detection in Incremental Learning Models",David Munechika (Rochester Institute of Technology); Ryne P Roady (Rochester Institute of Technology); Christopher Kanan (Rochester Institute of Technology)
- "Know thy Enemy: Invasive Species Detection in High-Resolution Imagery", Manoj Acharya (Rochester Institute of Technology)
- "Video Fingerprinting in Tor",Saidur Rahman (Rochester Institute of Technology); Nate J Mathews (Rochester Institute of Technology); Matthew Wright (Rochester Institute of Technology)
- "Chinese Sign language Interpretation using Deep learning techniques", Tejaswini Ananthanarayana (Rochester Institute of Technology)
- "Lifting restrictions on fluorescence microscopy through machine learning based super resolution and spectral unmixing",Tristan D McRae (University of Rochester); Yurong Gao (University of Rochester)
- 1:30-2:15pm, Keynote: David Doermann
- 2:15-2:30pm, snack break
- 2:30-3:30pm, PM Oral presentations:
- 2:30 pm - "Automatic Quantification of Facial Asymmetry using Facial Landmarks",Abu Md Niamul Taufique (Rochester Institute of Technology); Andreas Savakis (Rochester Institute of Technology); Jonathan Leckenby (University of Rochester Medical)
- 2:45 pm - "Fast Detection Based on Customized Complex Valued Convolutional Neural Network for Generalized Spatial Modulation Systems", Akram Marseet (University of Tripoli); Taiseer Elganim (University of Tripoli)
- 3 pm - "Synthetic data augmentation for improving low-resource ASR", Bao Thai (Rochester Institute of Technology); Robert Jimerson (Rochester Institute of Technology); Dominic Arcoraci (Rochester Institute of Technology); Emily Prud'hommeaux (Boston College); Raymond Ptucha (Rochester Institute of Technology)
- 3:15 pm - "Understanding Human and Predictive Moderation of Online Science Discourse", Elizabeth R Lucas (Rochester Institute of Technology); Cecilia Alm (Rochester Institute of Technology); Reynold Bailey (Rochester Institute of Technology)
- 3:30-4pm, Invited: Gonzalo Mateos Buckstein
- 4-4:30pm, Invited: Rob Phipps
- 2:30-4:30pm, RIT Research Computing by Sidney Pendelberry (parallel track – Room 2120)
- 4:30-4:45pm, Bryan Blakeslee 2018 Best paper presentation winner
- 4:45-5pm, Awards and wrapup
- Ziya Arnavut, SUNY Fredonia
- Nathan Cahill, Rochester Institute of Technology
- Edgar Bernal, University of Rochester
- Zhiyao Duan, University of Rochester
- Christopher Kanan, Rochester Institute of Technology
- Paul Lee, University of Rochester
- Cristian Linte, Rochester Institute of Technology
- Alexander Loui, Rochester Institute of Technology
- David Odgers, Odgers Imaging
- Raymond Ptucha, Rochester Institute of Technology
- Richard Zanibbi, Rochester Institute of Technology
Oral Presentation Instructions
All oral presentations will be 12 minutes long plus 2 minutes for questions. Presentors supply their own laptop with a VGA/HDMI connector. Morning/afternoon presentors should test their laptop on the display screen during the 8:15-8:45am or 12:45-1:45pm timeframes respectively. Papers whose first author is a student qualify for best paper award.Poster Presentation Instructions
All printed posters must be no larger than 40" wide x 48" tall. Poster stations will be available for both mounted and unmounted posters. (If you can bring a mounted poster, please do so.) Attachment materials will be provided. All posters must be displayed by 11am and removed by 5:30pm. There are no electrical outlets next to the poster displays. Posters whose first author is a student qualify for best poster award.If your department does not have poster printing capabilities, you can get posters printed at "the Hub Express" in the RIT Student Union, hubxpr@rit.edu, 585-475-3471. Color wide format inkjet is $7/sq.ft. Mounting (up to 30x40) is $4/sq.ft. (Contact the Hub Express if you have larger than 30x40)