Dr. Hoogs Presents Computer Vision Papers at IEEE Workshop
Dr. Anthony Hoogs, Kitware's Senior Director of Computer Vision, presented three papers at the 2014 IEEE Applied Imagery and Pattern Recognition (AIPR) Workshop. The workshop was held in Washington, DC, from October 14 to October 16, 2014.
For "Multi-Target Tracking in Video with Adaptive Integration of Appearance and Motion Models," Dr. Hoogs discussed a hybrid tracking architecture that utilizes motion detections to robustly initialize multiple tracks. The tracking architecture also uses a blended approach to integrate appearance-based trackers, provides a generalized API for interfacing such trackers, and adaptively uses motion detection or appearance match to update a track. The paper's authors are Arslan Basharat, Ilker Ersoy, Kannappan Palaniappan, Anthony Hoogs, and Gunasekaran Seetharaman.
For "KWiver: A Open-Source Cross-Platform Video Exploitation Framework," Dr. Hoogs introduced KWiver, a cross-platform video exploitation framework that Kitware has begun releasing as open source. The paper's authors include Keith Fieldhouse, Matthew J. Leotta, Arslan Basharat, Russell Blue, David Stoup, Charles Atkins, Linus Sherrill, Benjamin Boeckel, Paul Tunison, Jacob Becker, Matthew Dawkins, Matthew Woehlke, Roderic Collins, Matt Turek, and Anthony Hoogs.
Dr. Hoogs also presented "Towards visual analysis of unconstrained images in social forums: Studies on concept detection and personalized economy of images in social networks." The paper's authors are Sangmin Oh, Eric Smith, Yiliang Xu, and Anthony Hoogs.
Kitware Works with NVIDIA to Drive HPC Visualization
Kitware announced that it is working with NVIDIA to develop several powerful enhancements for the field of high-performance computing (HPC) visualization.
Among these is a new Visualization Toolkit (VTK) rendering backend, which targets graphics processing units (GPUs) and takes advantage of the flexible programmable pipelines available in modern systems. This development has already resulted in significant improvements in rendering performance, especially with large geometries (20 million+ triangles) that can now be rendered over 100 times faster.
“Through our collaboration with NVIDIA, we will enhance the open-source VTK and ParaView to better serve the HPC visualization community,” said Dr. Berk Geveci, Kitware’s Senior Director of Scientific Computing.
In addition, as part of the Department of Energy (DOE) Advanced Scientific Computing Research (ASCR) project “XVis: Visualization for the Extreme-Scale Scientific-Computation Ecosystem," Kitware, in collaboration with government labs and academia, is developing a unified visualization library, named VTK-m, as a single point to collaborate, contribute, and leverage massively threaded visualization algorithms. Kitware will be demonstrating VTK-m in NVIDIA GPU-accelerated applications in in situ mode within ParaView Catalyst.
NVIDIA will add future EGL support to its drivers, eliminating the need to install an X server for graphics output in ParaView. The companies are also exploring bringing OptiX parallel ray tracing and volume visualization capabilities into ParaView for better insight into simulations.
This collaboration will enable VTK and ParaView to fully benefit from the huge computational and graphics performance NVIDIA Tesla and Quadro architectures provide.
This material is based upon work supported by the U.S. Department of Energy, Office of Science, under Award Number DE-SC0012386.
Research reported in this publication was supported by the National Institute Of Biomedical Imaging And Bioengineering of the National Institutes of Health under Award Number R01EB014955. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
SC14 Showcase Highlights ParaView and VTK, Among Other Technology
Kitware exhibited recent work in HPC and visualization at Supercomputing 2014 (SC14) in New Orleans, LA. SC14 is one of the premier conferences in the scientific computing field.
As part of its ParaView Showcase, Kitware hosted presentations from renowned collaborators at its booth throughout the conference, where they discussed how they are leveraging ParaView, VTK, and other technology in their work. Featured presenters included Alejandro Ribes (Électricité de France), Sean Ziegler (Engility Corporation), James Ahrens (Los Alamos National Laboratory), Kenneth Moreland (Sandia National Laboratories), Paul Navratil (Texas Advanced Computing Center), Joseph Insley (Argonne National Laboratory), Jim Jeffers (Intel), and Alan Scott (Sandia National Laboratories).
Kitware was also involved in this year's event in a number of tutorials. On November 16, 2014, a team presented the tutorial "Large Scale Visualization with ParaView." This tutorial showcased the architecture of ParaView and the fundamentals of parallel visualization.
On the following day, Kitware and its collaborators presented the tutorial "In Situ Data Analysis and Visualization with ParaView Catalyst." This tutorial examined the architecture of ParaView Catalyst and the fundamentals of in situ data analysis and visualization. In addition, a group of presenters discussed "An Image-Based Approach to Extreme Scale In Situ Visualization and Analysis" in the Big Data Analysis session.
In conjunction with SC14, the Second Workshop on Sustainable Software for Science: Practice and Experiences (WSSSPE2) was held on November 16, 2014. At the workshop, Marcus D. Hanwell provided an overview of a collaborative paper, "Sustainable Software Ecosystems: Software Engineers, Domain Scientists, and Engineers Collaborating for Science." This paper focuses on the significance of collaborations when developing advanced scientific software, where sustainability, reproducibility, and extensibility are important.
Kitware also presented two papers at PyHPC 2014: "Python Enabled ParaViewWeb for HPC Analysis and Visualization" and "Scientific data analysis and visualization at scale in VTK/ParaView with NumPy."
Furthermore, members of Kitware participated in invited talks. On November 16, 2014, Sébastien Jourdain presented a talk on ParaView at the HPC Developer Showcase. Then, on November 19, 2014, Robert Maynard explored "VTK-M: Uniting GPU Acceleration Successes" at the NVIDIA booth.
Also on November 19, members of Kitware, as well as several of its collaborators, highlighted "In Situ MPAS-Ocean Image-based Visualization" at the Visualization and Data Analytics Showcase. This video discusses a novel framework for highly interactive, image-based in situ visualization and analysis that promotes exploration.
Kitware Features Apps Developed with Qt Framework at Developer Event
Qt Developer Days 2014 was held in Burlingame, CA, and featured presentations from experts in the Qt community. Throughout the three-day conference, Sankhesh Jhaveri showcased a variety of applications developed using the Qt framework, including ParaView, 3D Slicer, Slicer Extensions, VesselView, Bender, CMake, Avogadro, and VIVIA for Wide Area Motion Imagery (WAMI). Kitware has been one of the leading development teams for many of these cross-platform, open-source projects.
Bender is a free, open-source application for repositioning voxelized anatomical models.
Avogadro is a chemical editor and visualization application.
Dr. Turek Discusses Novel Framework at MSS National Symposium
Dr. Matt Turek, Kitware's Assistant Director of Computer Vision, presented two papers at the MSS National Symposium on Sensor and Data Fusion, which was held from October 28 to October 31, 2014, in Springfield, VA.
For "Interactive, Content-Based Exploration of Large Visual Archives through Feature Set Fusion," Dr. Turek discussed a novel framework Kitware is developing for visual data search and exploration for social multimedia. The framework combines computer vision, data fusion, and graph-based interactive visualization applied to content commonly found on the web, such as Youtube videos. The paper's authors are Sangmin Oh, Jeff Baumes, and Anthony Hoogs.
For "Probabilistic Sub-Graph Matching for Video and Text Fusion," Dr. Turek detailed Kitware's work in associating textual descriptions of entities and activities with automatically extracted visual entities (e.g., tracks) and activities (e.g., "person walking"), while also inferring the presence of missing details. The paper's authors are Eran Swears, Arslan Basharat, Anthony Hoogs, and Erik Blasch.
Dr. Enquobahrie Sheds Light on Need for Community Infrastructure in mHealth
On October 29, 2014, Dr. Andinet Enquobahrie participated in the National NSF workshop on Computing Challenges in Future Mobile Health (mHealth) Systems, which was held at the National Institutes of Health in Bethesda, Maryland. During the presentation, Dr. Enquobahrie shared Kitware's experience in developing software and community building tools for the medical imaging and EHR communities.
Dr. Enquobahrie discussed with the workshop's participants that a publicly accessible mHealth database and a robust community infrastructure will foster collaborations between mobile health researchers and computational scientists. In order to create a robust infrastructure that is sustainable and effective, however, key technical considerations must be taken into account during the design process. Such considerations include user-friendly and flexible access, the capability to handle heterogeneous data, a unified API, and standardized baseline algorithms.
To build a community around the database, it is important to find mutually beneficial incentives for collaboration between public health researchers, computer scientists, computational scientists, clinicians, and healthcare administrators. It is also critical to build trust among the community's members and to develop a common purpose.
Other key considerations discussed during the workshop were the benefits of a strong collaborative multidisciplinary research community, the major challenges in building a strong community, the best practices to initiate and nurture a vibrant and active community, and the role of open science (open data, open model, and open-source software).
The overall takeaway from Dr. Enquobahrie's presentation is that a well-designed publicly accessible database and collaboration infrastructure will encourage robust collaboration among researchers and lead to successful data-driven healthcare solutions.
Kitware and TACC Collaborate to Become Intel® Parallel Computing Center
Kitware has recently become an Intel® Parallel Computing Center in collaboration with Texas Advanced Computing Center (TACC). This team will focus its efforts on supporting the Intel® Xeon Phi™ Coprocessor in ParaView and in the Visualization Toolkit (VTK).
Today, HPC is widely recognized as a strategic capability of fundamental importance to the economic strength and national security of the United States and other developed countries, applicable to areas ranging from engineering simulation to medical computing to climate modeling. However, it can be difficult to deploy HPC at scale due to the complexity of the programming model and the lack of reference implementations with which to educate and accelerate the creation of new software solutions.
To address these challenges, the team of collaborators will work to add support for the Intel® Xeon Phi™ Coprocessor in VTK/ParaView. Adding this capability to these well-established and widely adopted open-source projects will not only provide easily deployable scalable solutions, but will also serve as a working example for software developers to study and extend.
“The opportunity to become an Intel® Parallel Computing Center will foster a collaboration between Kitware, TACC, and Intel that will enhance ParaView to run more efficiently on the Intel® Xeon Phi™ Coprocessor platform,” Dr. Berk Geveci, Kitware’s Principal Investigator for the project and Senior Director of Scientific Computing, said. “We are honored to be part of such a critical effort in HPC.”
For the project, Kitware and TACC will optimize VTK/ParaView to support Phi’s wide vector processing capability and to take advantage of the large number of computing cores. The team will develop support for both coarse and fine-grained parallelism and will release open-source software to demonstrate the use of these capabilities in a variety of scientific computing applications. By leveraging the multi-core and vector capabilities of the Intel® Xeon Phi™ coprocessors, the team will demonstrate a significant increase in the performance of VTK/ParaView.
Kitware’s scientific computing team will collaborate with TACC under the leadership of its Principal Investigator, Dr. Paul A. Navrátil. Dr. Navrátil is a Research Associate and the Manager of TACC’s Scalable Visualization Technologies group. He is an expert in high-performance visualization technologies and advanced rendering techniques, including algorithms for large-scale distributed-memory ray tracing.
Kitware Fuels Scientific Discovery with ASCR Awards
Kitware recently announced its collaboration on three Advanced Scientific Computing Research (ASCR) awards from the Department of Energy (DOE).
The awards are part of the ASCR Computer Science program, which supports research that enables computing at extreme scales and the understanding of extreme-scale data from both simulations and experiments. Recently, the program announced nine new awards under Scientific Data Management, Analysis & Visualization at Extreme Scale 2.
“We are honored to be selected for three of the ASCR awards,” Dr. Berk Geveci, Kitware’s Principal Investigator on two of the projects and the company’s Senior Director of Scientific Computing, said. “The Computer Science program is important because it addresses key areas in the field of scientific computing, including high-performance computing (HPC) and extreme-scale simulations.”
For the first project, “XVis: Visualization for the Extreme-Scale Scientific-Computation Ecosystem,” Kitware will collaborate with Sandia National Laboratories, under the leadership of Principal Investigator Ken Moreland; Oak Ridge National Laboratory; Los Alamos National Laboratory; University of California Davis; and University of Oregon.
The team will build a unified visualization library, named VTK-m, as a single point to collaborate, contribute, and leverage massively threaded visualization algorithms. The motivation behind the project is the shift in HPC toward the increasing use of accelerators and other processor technologies that use greater concurrent threading to overcome physical limitations of power and latency.
Building on top of this foundation, the XVis project will research techniques for flyweight in situ components, advanced data models, new domain challenges, and post hoc interaction techniques.
Secondly, on the “Scalable Analysis Methods and In Situ Infrastructure for Extreme Scale Knowledge Discovery” project, Kitware will collaborate with Lawrence Berkeley National Laboratory, under the leadership of PI Wes Bethel; Argonne National Laboratory; Georgia Institute of Technology; and JMSI, Inc.
The project will address a set of research challenges for enabling scientific knowledge discovery within the context of in situ processing at extreme-scale concurrency. The work is motivated by a widening gap between FLoating-point Operations Per Second and input/output (I/O) capacity, which will make full-resolution, I/O-intensive post hoc analysis prohibitively expensive, if not impossible. The team will focus on developing new algorithms for analysis and visualization that are suitable for use in an in situ context aimed specifically at enabling scientific knowledge discovery in several exemplar application areas of importance to the DOE.
Lastly, Kitware will collaborate with Los Alamos National Laboratory, under the leadership of Principal Investigator Pat McCormick; Sandia National Laboratories; University of Utah; and Stanford University on the project “A Unified Data-Driven Approach for Programming In Situ Analysis and Visualization.”
The overarching goal of the project is the study of a unified data-driven approach for programming applications, as well as in situ analysis and visualization. In particular, the team will focus on the interplay between data-centric programming model requirements at extreme scale and the overall impact of those requirements on the design, capabilities, flexibility, and implementation details for both applications and the supporting in situ infrastructure.
“We are looking forward to collaborating with world-renowned organizations on projects that will significantly advance scientific discovery,” Geveci said.
This material is based upon work supported by the U.S. Department of Energy, Office of Science, under Award Numbers DE-SC0012387, DE-SC0012388, and DE-SC0012386.
Virtual Surgery System Aims to Help Surgeons Treat Craniosynostosis
Kitware is developing software for treatment planning and evaluation of craniosynostosis through robust, quantitative, and reproducible methods that assess cranial shape.
Craniosynostosis is the premature fusion of cranial sutures. It occurs in approximately one in 2,000 live births and results in cranial malformation that can lead to elevated intra-cranial pressure and brain growth impairment. It can also lead to developmental deficiency.
While the most common treatment option for craniosynostosis is surgery, today’s surgical treatment planning is mostly qualitative, subjective, and irreproducible. In addition, although virtual planning has been successfully introduced in niche areas of craniofacial surgery, such as corrective jaw surgery applications, there exists a need for clinical tools that provide accurate and reproducible evaluation of cranial morphology to guide cranial vault remodeling.
“To cover the gap in current clinical practice, we will develop personalized technology for preoperative planning for infants with craniosynostosis,” Dr. Andinet Enquobahrie, Kitware’s Principal Investigator for this project and Assistant Director of Medical Computing, said. “The technology will help decrease operative time and blood loss to reduce perioperative morbidity and facilitate an optimized and more durable long-term outcome.”
The goal of the Phase I Small Business Technology Transfer (STTR) project funded by the National Institutes of Health is to design, develop, and validate a virtual surgery system for optimal treatment planning for cranial remodeling. To accomplish this goal, the team will create a normative multi-atlas database that captures a wide breadth of normal variations on cranial shape and patient ages. The team will also develop image processing, statistical shape analysis, and biomedical modeling algorithms to identify desirable post-operative skull shapes.
The project is a collaboration between Kitware and Children’s National Health System in Washington, DC. Children's National is an internationally recognized pediatric hospital, with over 140 years of experience, and a center of excellence in the treatment of birth defects and the clinical management of craniosynostosis. Children’s National is also home to the Sheikh Zayed Institute for Pediatric Surgical Innovation, which is devoted to advancing the frontiers of pediatric surgery.
“The impact of this technology is reduced perioperative morbidity and lower treatment costs,” Dr. Marius Linguraru, the Principal Investigator for the project and a Principal Investigator and Director of the Quantitative Imaging Group of the Sheikh Zayed Institute for Pediatric Surgical Innovation, said. “The technology will also enable the precise, quantitative comparison of measurements before and after cranial vault reconstruction to determine the efficacy and durability of specific reconstructive techniques.”
The project will build upon Kitware’s extensive experience in the field of surgical guidance. As the primary developer of Image Guided Surgery Toolkit (IGSTK), Insight Registration and Segmentation Toolkit (ITK), and 3D Slicer, Kitware has demonstrated leadership in the creation, integration, and support of state-of-the-art medical data visualization and analysis technology.
Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute Of Child Health & Human Development of the National Institutes of Health under Award Number R41HD081712. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Kitware Prints New Books On Demand
Kitware is pleased to announce that the newest versions of the ITK, ParaView, and CMake books are now available through Amazon CreateSpace. In an effort to best contribute to the open-source communities in which Kitware participates, the books are now being published as print-on-demand offerings, enabling Kitware to provide an updated book with each software release.
The ParaView Guide is updated for the 4.3 release, and Mastering CMake is updated for the 3.1 release. Additionally, The ITK Software Guide has been completely rewritten and split into two books: 1) Introduction and Development Guidelines and 2) Design and Functionality.
More information, including links for ordering the books, is available on the Kitware website. Check out the updated books now!
Sujin joined Kitware as a Research and Development Engineer on the Scientific Computing team. Sujin received his B.E. from the University of Pune and his M.S. in Computing – Graphics and Visualization from the University of Utah. Prior to joining Kitware, Sujin was a Research Assistant at the Scientific Computing and Imaging Institute. There, he developed a scalable, parallel, out-of-core solver for gradient domain processing of large, gigapixel-sized images using multithreading, GPGPU, and MPI. Sujin also developed a scalable, parallel, out-of-core technique for computing seams for gigapixel-sized panoramas using multithreading and MPI.
Javier joined Kitware as an Assistant Office Administrator. Javier attends Hudson Valley Community College, where he is studying repiratory care. Prior to joining Kitware, Javier was a Customer Service Representative at Pioneer Bank and a Hyperbaric Technician at Mobile Hyperbaric Centers. Javier was also a Teacher Associate at the Center for Discovery in Harris, NY, where he assisted in teaching children with developmental disabilities skills for achieving higher levels of independence.
Dženan joined Kitware as a Research and Development Engineer on the Medical Computing team. Dženan has defended his thesis and is currently finishing up his Ph.D. dissertation at the University of Siegen. He is studying vertebral body segmentation and diagnosis in magnetic resonance images. Prior to joining Kitware, Dženan was a System Engineer at BSTelecom, where he designed and implemented a web interface for a database-backed inter-telecom billing system. Dženan also has experience as a developer at aNET. As such, he designed a database, a GUI, and a web interface for the technical examination and registration of vehicles.
Naval Future Force Science And Technology (S&T) Expo
February 4 to February 5, 2015, Washington, DC
The expo is sponsored by the Office of Naval Research (ONR). Topics to be addressed at the event include objectives of the revised Naval S&T Strategy and new research opportunities within the Navy and Marine Corps community. The status of key programs will also be highlighted.
Presentation on Open Source at the College of Charleston
February 6, 2015, Charleston, SC
Stephen Aylward and Jean-Christophe Fillion-Robin will present "Accelerating Medical Visualization Research via Open-Source Platforms and Practices." This talk will introduce the tools that enable the creation of effective open-source platforms, discuss the costs and benefits of creating and using open-source platforms, and identify the trends of open-source visualization in medicine.
Specific examples will be given throughout the talk, drawn from ongoing research and development projects at Kitware, including recent enhancements to 3D Slicer and research into USB-based ultrasound probes and automated ultrasound analysis software for in-field brain and abdominal trauma assessment.
The talk will conclude with a live demonstration that highlights innovations in 3D Slicer for interactive disease visualization and treatment planning.
Rensselaer Polytechnic Institute Spring 2015 Career Fair
February 11, 2015, Troy, NY
Members of Kitware will be recruiting candidates for job and internship opportunities at the career fair. For a list of available positions, please visit http://jobs.kitware.com/opportunities.html.
SPIE. Medical Imaging
February 21 to February 26, 2015, Orlando, FL
The conference will consist of sessions, keynote addresses, panels, workshops, poster presentations, and interactive demos. Topics of discussion will include image processing, computer-aided diagnosis, image-guided procedures, ultrasonic imaging and tomography, and digital pathology.
For more information, please visit the conference's webiste at http://spie.org/medical-imaging.xml.
SIAM Conference on Computational Science and Engineering
March 14 to March 18, 2015, Salt Lake City, UT
Kitware has been actively involved in planning this event, as Patrick O’Leary is on the Organizing Committee. Kitware will also have a strong presence at the event, beginning with Will Schroeder's keynote address. In addition, members of Kitware will present several minisymposiums, including "Software Quality with the Open Source Tools CMake, CDash, CTest” and "Computational Model Builder and ParaView Catalyst: Empowering HPC Workflows.”
Kitware will also present a minisymposium titled "Software Process for a CASL Sustainable Simulation Software Solution” and a minitutorial titled "Python Visual Analytics for Big Data." Furthermore, Kitware will attend the Career Fair, which will take place on March 14.
For more information, please visit the conference's website at http://www.siam.org/meetings/cse15/index.php.
GPU Technology Conference
March 17 to March 20, 2015, San Jose, CA
The conference will host scientists, programmers, researchers, and other professionals. It will feature keynote addresses, exhibits, receptions, sessions, hangouts, and labs.
At the conference, Robert Maynard and Marcus D. Hanwell will present the paper "Visualization Toolkit: Faster, Better, Open! Scientific Rendering and Compute."
For more information, please visit the conference's website at http://www.gputechconf.com/.
IEEE Virtual Reality 2015
March 23 to March 27, 2015, Arles, Camargue – Provence, France
Patrick O'Leary is an organizer of the Third Workshop on Immersive Volumetric Interaction (WIVI 2015). According to the conference's website, the goal of the workshop is to unite researchers, developers, and users from several communities in Virtual Environments, 3D User Interfaces, and Immersive Visualizations. At the workshop, attendees will talk about issues and present ideas in regards to the challenge of interacting with volumetric data in immersive visualizations. Multiple objects defined as a volume in virtual environments will also be discussed.
For more information, please visit http://ieeevr.org/2015/.
Kitware is seeking talented, motivated, and creative individuals to fill open positions. Kitware has an immediate need for software developers and researchers who will work on cutting-edge research and join us in our mission to develop and deliver state-of-the-art software products and services using advanced software quality methods and technologies.
The impact of your research at Kitware will extend far beyond our organization, as our open-source business model will allow you to become part of the worldwide communities that surround our projects. Kitware employees enjoy a collaborative work environment that empowers them to pursue new opportunities and to challenge the status quo through novel ideas.
In addition to providing an excellent workplace, Kitware offers comprehensive benefits including: flexible hours; a computer hardware budget; health, vision, dental, and life insurance; short- and long-term disability; visa processing; a generous compensation plan; and free drinks and snacks.
For additional information, please visit our employment website at jobs.kitware.com. Interested applicants are encouraged to submit their resumes and cover letters through our online portal.
Kitware internships provide current college students with the opportunity to gain hands-on experience working with leaders in their fields on cutting-edge problems. Our interns assist in developing foundational research and leading-edge technology across five business areas: HPC & visualization, computer vision, medical computing, data and analytics, and quality software process. We not only offer our interns a challenging work environment, but we also provide them with the opportunity to attend advanced software training.