Kitware at CVPR 2021
Watch our capabilities video
Schedule of Events
Kitware is actively participating in the fully virtual CVPR 2021 conference. We are helping to organize and present at the full-day workshop on media forensics, and we are once again proud to sponsor the EarthVision workshop. And finally, we were honored to have members of Kitware’s Computer Vision and Data and Analytics Teams develop the visualizations on CVPR’s website to accommodate virtual attendees.
Join us virtually for these featured events:
Saturday, June 19th
EarthVision: Large Scale Computer Vision for Remote Sensing Imagery
Earth observation and remote sensing depend on computer vision, machine learning, and signal/image processing. Together, they provide large-scale, homogeneous information about processes occurring at the Earth’s surface, which can largely impact human society, economy, industry, and the planet. This workshop aims to raise awareness of this highly challenging and quickly evolving field of research and to foster collaboration between the computer vision and Earth observation communities. Click for more information about the workshop.
Matt Leotta, technical leader at Kitware, is part of the technical committee for the workshop.
Saturday, June 19th
Workshop on Media Forensics
Media forensics has recently become a topic of interest within the computer vision community as a result of the increasing prevalence of fabricated media spreading misinformation. And now, with the emergence of more sophisticated machine learning and computer vision techniques, media forensics has become a broad and prominent area of research. With the emergence of more sophisticated machine learning and computer vision techniques, these experts will discuss the importance of advancing media forensics. Click for more information about the workshop.
Scott McCloskey, assistant director of computer vision at Kitware, is a member of the organizing committee, and Arslan Basharat, assistant director of computer vision at Kitware, is part of the technical committee for the workshop.
Wednesday, June 23rd
Paper Presentation: Discovering Hidden Physics Behind Transport Dynamics
Transport processes are ubiquitous. They are, for example, at the heart of optical flow approaches; or of perfusion imaging, where blood transport is assessed, most commonly by injecting a tracer. An advection-diffusion equation is widely used to describe these transport phenomena. Our goal is to estimate the underlying physics of advection-diffusion equations, expressed as velocity and diffusion tensor fields. We propose a learning framework (YETI) building on an auto-encoder structure between 2D and 3D image time-series, which incorporates the advection-diffusion model. To help with identifiability, we develop an advection-diffusion simulator which allows pre-training of our model by supervised learning using the velocity and diffusion tensor fields. Instead of directly learning these velocity and diffusion tensor fields, we introduce representations that assure incompressible flow and symmetric positive semi-definite diffusion fields and demonstrate the additional benefits of these representations on improving estimation accuracy. We further use transfer learning to apply YETI on a public brain magnetic resonance (MR) perfusion dataset of stroke patients and show its ability to successfully distinguish stroke lesions from normal brain regions via the estimated velocity and diffusion tensor fields.
Looking for expertise in deep learning for computer vision?
Are you interested in joining our Computer Vision Team?
We are committed to furthering research and development in computer vision.
Our Deep Learning Open Source Platforms
VIAME is an open source, do-it-yourself AI system for analyzing imagery and video for general use, with specialized tools for the marine environment.
TeleSculptor is a cross-platform application for aerial photogrammetry.
SMQTK is an open source toolkit for exploring large archives of image and video data that enables users to easily and dynamically train custom object classifiers for retrieval.
KWIVER is an open source toolkit that solves challenging image and video analysis problems.