Joseph L. Mundy
Dr. Mundy joined General Electric’s Research and Development Center (CRD) in 1963. His early projects at CRD include: High power microwave tube design, superconductive computer memory devices, the design of high density integrated circuit associative memory arrays, and the application of transform coding to image data compression. He is the co-inventor of varactor bootstrapping, a key technique still widely used today in the design of CMOS integrated circuits. From 1972 until 2002, Dr. Mundy led a group involved in the research and development of image understanding and computer vision systems. In the early 1970’s his group developed one of the first major applications of computer vision to industrial inspection. A system was developed to inspect incandescent lamp filaments at the rate of 15 parts/sec. and achieved classification performance of less than one error per thousand. The system operated in production for many years. His more recent research themes at CRD included: industrial photogrammetry for machine control, theory of geometric invariance, change detection in satellite imagery and CT classification of lung cancer lesions. In 1988, Dr. Mundy was named a Coolidge Fellow, GE’s highest technical honor. He applied the fellowship to a sabbatical at Oxford University, working with Sir Michael Brady and Prof. Andrew Zisserman on applications of invariant theory to computer vision. This work lead to the Marr Prize award in 1993. In 2002 Dr. Mundy joined the School of Engineering at Brown as Prof. of Engineering (research). His research at Brown, under sponsorship of DARPA and NGA, includes video and image analysis, with emphasis on change detection and 3-d volumetric modeling. In 2011, Dr. Mundy co-founded Vision Systems Inc. and is President and CEO. At VSI, Dr. Mundy has managed the DARPA Tailwind and Visual Media Reasoning projects that are aimed at aerial video processing and scene analysis. Dr. Mundy also provides general project management of current VSI efforts in 3-d modeling from satellite imagery, geo-location of ground-level imagery and facial recognition. Dr. Mundy received his B.E.E.(1963) and M.Eng.(1966) and Ph.D.(1969) from Rensselaer Polytechnic Institute. He has published over 100 papers and articles in computer vision and solid state electronics.
Dr. Crispell received his BS from Northeastern University in 2003 and Ph.D. from Brown University in 2010 where he focused on probabilistic 3-d reconstruction algorithms using volumetric geometry representations. Dan is a co-founder of Vision Systems and has published several research papers in the fields of 3-d reconstruction, image and video registration, and change detection. He has developed and applied 3-d reconstruction and automatic change detection processing techniques to overhead imagery and developed automatic fusion and exploitation algorithms for LIDAR, image, and video data. He has implemented a belief propagation framework for inferring 3-d scene representations of single images based on fusion of outputs from a large set of independent off-the-shelf computer vision algorithms as part of the DARPA Visual Media Reasoning (VMR) program. Current areas of investigation also include the use of 3-d models to aid in the human face recognition task.
Dr. Dong received his PhD degree in 2012 from the School of Engineering, Brown University, along with MSc in applied mathematics and engineering (2009, 2010). He has since been focusing on areas including 3-d scene modeling from satellite images, image registration, stereo matching, region classification in multi-spectral satellite imagery, change detection, semantic web, AWS cloud approaches, and natural language processing. He has published a series of research publications in the field of computer vision and material science. He served as research scientist of IARPA FINDER program and an AFRL SBIR program to construct large scale high resolution Digital Elevation Model (DEM), geo-correct various remote sensing data and geo-locating ground-level imagery using various GIS reference data and satellite images. Yi is currently working as the principal investigator of a DARPA SBIR program to develop and demonstrate methods that leverage satellite imagery and semantic web data for use in various global scale applications.
Scott Richardson finished his Masters degree in 2007 from the University of Colorado (CU) in Computer Science with an emphasis in Machine Learning. While at CU, he competed in the DARPA Learning Applied to Ground Robotics (LAGR) challenge in which he implemented path-planning algorithms for navigation through unstructured environments. Scott worked for Lockheed Martin as a Research Scientist before joining VSI in 2012. Scott has experience developing 3D scene modeling applications and is a member of the VXL developer team. As part of the DARPA Visual Media Reasoning (VMR) program he designed a Factor Graph Toolkit based on the GraphLab graph-parallel processing framework to perform distributed belief propagation. He is currently investigating Generative Adversarial Networks and their application to face detection.
Andy Neff received his BS (2004) and MS (2007) in Electrical Engineering from Clemson University with an emphasis on Intelligence Systems. He worked for Lockheed Martin for almost 7 years as a Research Scientist, working as an image processing algorithm developer and software integrator. His recent work includes VoxelGlobe, a web site interface for VSI algorithms powered by Django.
Dr. Biris finished his PhD in 2015 from the School of Engineering from Brown University, along with his ScB in Computer Engineering in 2009. His dissertation focused on the compression of 2-d and 3-d dynamic scenes based on motion estimation. He successfully adapted classical optical flow algorithms to work on 3-d time-varying probabilistic volumetric models by leveraging 3-d surface and appearance information. During his first years of his PhD, Octavian has developed an algorithm for surveillance video compression based on background modeling and the JPEG 2000 standard. His research interests also pursued new motion estimation methods based on recasting Schroedinger's equation in the context of computer vision. He was also the teaching assistant of the graduate level course on scientific programming in C++ for four years in a row, where he contributed to multiple lectures on modern C++11 features and development, Python C bindings and GPU computing. Octavian is currently working on combining 3-d human face modeling techniques with machine learning and data-driven methods in order to solve face detection and recognition problems.
Hasnain received his MS degree in Computer Science from Brown University in 2015. His studies at Brown were focused on computer vision and systems. Hasnain has worked on using deep learning for object detection in satellite images and his most recent work includes deep learning for facial recognition using 3D face models.
Elaina Lucas received her BS and MA in Management from Salve Regina University. Elaina provides leadership support of the daily operations for the Vision Systems team. Her organizational abilities contribute to the success of an array of tasks which encompass facilities management and supplies, employment record keeping, project contracts, corporate documents, expense reports, and general human resource issues. Elaina worked for Rhode Island School of Design as the Assistant Registrar for 7 years and Academic Facilities Manager for 12 years before she joined the VSI team in 2015.
Dr. Gilliam earned his Ph.D. in Electrical Engineering from the University of Virginia in 2008, developing innovative computer vision algorithms for state-of-the-art biomedical imaging modalities. He has since worked on a wide range of challenging computer vision problems across the spectrum of university, commercial, defense, and intelligence customers. Dr. Gilliam an expert in geospatial image & video analysis and understanding, specializing in the unique needs of the defense and intelligence communities.
Max graduated from Boston University in 2016 with a bachelor's degree in Computer Science. His two main areas of focus were distributed systems and Artificial Intelligence. Developing for mobile platforms, Max was part of a small team winning Best App Design in the 2016 Global App Initiative. His previous research experience explored the use of 3-d in robust object tracking and occlusion situations. Max joined VSI full time in June 2016, and is now working on an online, real-time, and highly scalable Structure from Motion (SfM) solution.
Dr. Sorensen received his MA and PhD from the College of Engineering at the University of Delaware in 2017. He earned a BS in computer science and BA in mathematics from Rowan University in 2010. His major areas of study included image based 3D reconstruction, thermal imaging and virtual reality application development. As part of his thesis work Scott was involved with the development and deployment of 3D camera systems in polar environments and participated in multiple field expeditions to the Arctic. He has published in a diverse set of computer vision topics, including material classification, 3D reconstruction, biomedical figure classification, and virtual reality visualization of scientific data. His current work includes the use of satellite imagery for 3D reconstruction and high level understanding.
Dr. Robert Wagner received his Ph.D. in Cognitive and Neural Systems from Boston University in 2004 and his B.Sc. in Computer Science from Purdue University in 1989. His research focused on both biological vision as well as computer vision to develop novel algorithms for robot navigation and control. While earning his doctorate, he also worked as a consultant to the Naval Research Lab where he developed a face detection and recognition system so that robots would turn towards and listen to spoken commands from a known person. Since 2004, Dr. Wagner has developed computer vision systems using both classical computer vision and deep learning for a variety of applications including: video fidelity enhancement, video surveillance, and large-scale visual search and retrieval. Prior to joining VSI, he focused his efforts on real-time object detection and tracking in the video surveillance domain. As part of this effort, he developed a deep learning data collection and image generation workflow for training a CNN-based weapon detector that is agnostic to background, pose, scale, and blur. He also created a test harness to compare different CNN models on a curated video corpus from in-house and YouTube sourced video. The best model was found by finding the model with the minimum inference time that provided high recall with low false alert rates. He is currently investigating the production of large-scale models of the surface of the earth.
Director of Finance and Administration
Denise received a BS in Accounting from Bryant College in 2003. Prior to joining VSI in 2016, she had 10 years’ experience as an auditor in State Government, and has worked in several bookkeeping/accounting positions in the private sector. Denise is responsible for the financial and administrative functions of the company.