
IEEE Applied Imagery Pattern Recognition Workshop Dedicated to facilitating the interchange of ideas between government, industry, and academia 45th AIPR Workshop Imaging and Artificial Intelligence: Intersection and Synergy Cosmos Club, Washington, D.C. October 18-20, 2016 AIPR 2016 is sponsored by the IEEE Computer Society Technical Committee on Pattern Analysis and Machine Intelligence with generous support from Cybernet Systems Corporation, Descartes Labs, Inc., and The Charles Stark Draper Laboratory, Inc. AIPR Workshop Committee Chairman Peter Doucette, U.S. Geological Survey Deputy Chair Daniela Moody, Los Alamos National Laboratory Program Chairs Murray H. Loew, George Washington University Kannappan Palaniappan, University of Missouri at Columbia Secretary Carlos Maraviglia, Naval Research Laboratory Treasurer Al Williams, Self-employed Engineer Local Arrangements Donald J. Gerson, Gerson Photography Publicity Peter Costianes, Air Force Research Laboratory, Emeritus Webmaster Charles J. Cohen, Cybernet External Support John Irvine, Draper Laboratory Christoph Borel-Donohue, Army Research Laboratory Registration Chair Jeff Kretsch, Raytheon BBN Technologies Proceedings Chair Franklin Tanner, Hadron Industries Student Paper Award Chair Paul Cerkez, DCS Corp. AIPR Committee Members: Jim Aanstoos, Mississippi State University Robert Mericsko, Booz Allen Hamilton Eamon Barrett, Retired Keith Monson, FBI Bob Bonneau, AFOSR Carsten Oertel, MITRE Filiz Bunyak, University of Missouri- William Oliver, Brody School of Medicine Columbia Kannappan Palaniappan, University of John Caulfield, Cyan Systems Industries Missouri-Columbia Charles Cohen, Cybernet James Pearson, Remote Sensing Consultant Peter Costiannes, AFRL, Emeritus Surya Prasath, University of Missouri-Columbia Paul Cerkez, DCS Corp. Amit Puntambekar, Intel Corporation Peter Doucette, U.S. Geological Survey Mike Pusateri, LSI Corporation Roger Gaborski, RIT Harvey Rhody, RIT Donald J Gerson, Gerson Photography David Schaefer, GMU Neelam Gupta, ARL Guna Seetharaman, NRL Mark Happel, Johns Hopkins University Karl Walli, AFIT John Irvine, Draper Lab Elmer "Al" Williams, Self-Employed Engineer Steve Israel, Integrity Applications/ONR Alan Schaum, NRL Michael D. Kelly, IKCS Jeff Kretsch, Raytheon BBN Technologies Emeritus Murray H. Loew, GWU Robert Haralick, City University of New York Carlos Maraviglia, NRL Joan Lurie, GCC Inc. Paul McCarley, AFRL Eglin AFB J. Michael Selander, MITRE 2016 AIPR Visionary Sponsors We are tackling the huge problem of teaching computers how to see. Our underlying technology, deep learning, models the way the human brain works. Though this was just science fiction even five years ago, advances in the speed of computers and the underlying software make it possible now. Further, because sensors can detect beyond the human visual spectrum, computers can see things that humans simply cannot. Our first project is building a living atlas of the world. We process petabytes of satellite imagery and extract daily insights that drive decision making in agriculture, finance, natural resources, public policy, and more. We work to create real-time global awareness to better understand changes over time and what is happening in our world right now. In the Press Deep-Learning, Image-Analysis Startup Descartes Labs Raises $3.3M After Spinning Out Of Los Alamos National Lab techcrunch.com – http://tcrn.ch/1bKgXIH After being incubated in a government lab for seven years, Descartes Labs — a deep-learning image analysis startup — has raised $3.3 million of its own. A Cloud-Free Satellite Map of Earth technologyreview.com – http://bit.ly/1p2yzXy Stitching together a live world map from many different satellite images lets algorithms keep an eye on the health of crops and problems like flooding. We’re Hiring! Our team includes cosmologists, physicists, mathematicians, engineers, computer scientists, and even a philosopher. We need people who are brilliant coders and stellar problem solvers - we welcome different backgrounds. We are headquartered in the fabulous mountains of Los Alamos, NM, and also have an office in San Francisco, CA. If this intrigues you, please reach out to us at [email protected]. We’d love to hear from you! 45th IEEE AIPR Workshop Program, October 2016 Tuesday October 18 Deep Learning & AI 8:00 AM Registration opens Note: For brevity, this schedule lists only the presenter of each paper; all authors and their affiliations are included with each abstract, indexed using the # in the schedule. Jason Matheny, Director, IARPA: Value of Machine Learning 9:00-9:45 AM for National Security AI and Image Processing I #20 Real-time detection and classification of traffic light signals; 9:45-10:05 AM Asaad Said, Intel #79 Machine vision algorithms for robust pallet engagement and 10:05-10:25 AM stacking; Charles Cohen, Cybernet Systems Corporation 10:25-10:40 AM Coffee Break Deep Learning for Image Processing I #6 SPARCNN: SPAtially Related Convolutional Neural Networks; 10:40-11:00 AM J.T. Turner, Knexus Research Corporation #47 Crossview convolutional networks; Nathan Jacobs, University of 11:00-11:20 AM Kentucky Trevor Darrell, EECS, University of California-Berkeley: 11:25-12:00 PM Perceptual representation learning across diverse modalities and domains 12:00-1:30 PM Lunch on your own Christopher Rigano, Office of Science and Technology, 1:30-2:15 PM National Institute of Justice: Applied Computer Science in Criminal Justice AI and Image Processing II #19 Trajectory based event classification from UAV videos and its 2:15-2:35 PM evaluation framework; Sung Chun Lee, Object Video, Inc. #56 Transfer learning for high resolution aerial image classification; 2:35-2:55 PM Yilong Liang, Rochester Institute of Technology #22 Visualization of high dimensional image features for 2:55-3:15 PM classification; Katie Rainey, SPAWAR Systems Center Pacific 3:15-3:30 PM Coffee Break 3:30-4:05 PM John Kaufhold, Deep Learning Analytics: Deep Learning Past, Present and Near Future Deep Learning for Image Processing II #60 Hierarchical temporal memory for visual pattern recognition; 4:05-4:25 PM Jianghao Shen, George Washington University #52 Hierarchical autoassociative polynomial network (hap net) for 4:25-4:45 PM robust face recognition; Theus Aspiras, University of Dayton #63 Convolutional neural network application to plant detection, 4:45-5:05 PM based on synthetic imagery; Larry Pearlstein, The College of New Jersey 5:10-5:30 PM Poster Preview 5:30-6:00 PM Cosmos Club Tour 6:00 PM Poster Session and Reception Tuesday Poster Session: Poster numbers, titles, and presenting author (full list of authors appears on the Abstract page) 2 CSANG: continuous scale anisotropic Gaussians for robust linear structure extraction; V. B. Surya Prasath, University of Missouri-Columbia 7 Seismic signal analysis using multi-scale/multi-resolution transformations; Millicent Thomas, Northwest University 9 Assessment of three-dimensional printing using visible light sensing and a CAD model; Jeremy Straub, University of North Dakota 17 Object identification for inventory management using convolutional neural network; Nishchal K. Verma, Indian Institute of Technology Kanpur, India 30 Multi-scale transformation based image super-resolution; Soundararajan Ezekiel, IUP 33 Keypoint density-based region proposal for fine-grained object detection and classification using regions with convolutional neural network features; JT Turner, Knexus Research Corporation 35 General purpose object counting algorithm using cascaded classifier and learning; Ricardo Fonseca, Universidad Tecnica Federico Santa Maria 44 Adding a selective multiscale and color feature to loft base line; Noor Al-Shakarji, University of Missouri Columbia 45 Auto-calibration of multi-projector systems on arbitrary shapes; Mahdi Abbaspour Tehrani, UC- Irvine 54 Boosted ringlet features for robust object tracking; Evan Krieger, University of Dayton 58 Video haze removal and Poisson blending based mini-mosaics for wide area motion imagery; Rumana Aktar, University of Missouri-Columbia 62 Deep Learning and Artificial Intelligence Completeness; Mihaela Quirk, Quirk & Quirk Consulting 66 Stream implementation of the flux tensor motion flow algorithm using Gstreamer and CUDA; Dardo Kleiner, CIPS, Corp. @ NRL 78 Enhanced GroupSAC: Efficient and Reliable RanSAC scheme in the presence of depth variation; Kyung-Min Han, University of Missouri Wednesday High-Performance Computing & Biomedical October 19 8:00 AM Registration opens Richard Linderman, SES, Office of the Assistant Secretary of Defense, 8:45-9:30 AM Research and Engineering: DoD Interests in Autonomy and Artificial Intelligence Scalable Image Processing #26 Scaling analytics for scientific images from experimental instruments; 9:30-9:50 AM Daniela Ushizima, Lawrence Berkeley National Laboratory #73 Computationally efficient scene categorization in complex dynamic 9:50-10:10 AM environments; Allison Mathis, Army Research Laboratory 10:10-10:30 #53 Uncertainty of image segmentation algorithms for forensic attribution of AM materials; Diane Oyen, Los Alamos National lab 10:30-10:45 Coffee Break AM Earth & Space Sensing Applications 10:45-11:05 #49 Sparse multi-image control for remotely sensed planetary imagery: the AM Apollo 15 metric camera; Jason Laura, United States Geological Survey 11:05-11:25 #57 On the utility of temporal data cubes for analyzing land dynamics; Peter AM Doucette, USGS 11:25-11:45 #14 Training and evaluating detection pipelines with connected components;
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages65 Page
-
File Size-