Jason Parham received his B.S. in computer science/mathematics in 2008 from Pepperdine University in Malibu, California. In addition to this degree, he holds an M.S. in computer science from Rensselaer Polytechnic Institute (RPI). At RPI, he wrote his Master’s thesis on the design and implementation of citizen science-powered photographic censuses of zebra and giraffe in Nairobi National Park in Kenya.
Alongside his full-time work at Kitware, Jason is pursuing his Ph.D. in computer science as a part-time student at RPI. His advisor in the computer vision research group at RPI is Dr. Charles Stewart. At the start of his employment at Kitware, Jason had completed five years of his Ph.D. program as a full-time graduate student. His current doctoral research focuses on object detection and classification. It uses deep learning on wildlife imagery to power photographic censusing.
Also related to wildlife imagery, Jason is a co-developer of Image-Based Ecological Information System (IBEIS). This image-analysis software is used to monitor animal populations in conservancies around Kenya. It integrates with Wildbook, an open-source platform for wildlife management.
Starting in 2015, Jason worked for three consecutive summers as a research intern on the computer vision team at Kitware. His internships focused on writing proposals and contributing to analytical software for aerial and satellite change detection, using fully convolutional neural networks (FCNNs) and generative autoencoder networks (GANs). His efforts resulted in the creation of the KitWare Convolutional Neural Network (KWCNN). This Python module enables easy and effective deep learning applications at Kitware such as the Visual Global Intelligence and Analytics Toolkit (VIGILANT).
Moving forward, Jason’s work at Kitware will focus on training deep learning algorithms and integrating them with aerial-based detection. He will also continue to develop KWCNN.