Research
About Our Work
VIVA focuses on developing cutting edge technology for analyzing image and video datasets - with a wide variety of applications ranging from biomedical image analysis, human activity recognition and other computer vision problems.
The Virginia Image and Video Analysis (VIVA) Lab, affiliated with the Charles L. Brown Department of Electrical and Computer Engineering at the University of Virginia is led by Professor Scott T. Acton. Our lab focusses on developing principled algorithms for biomedical image analysis, human activity recognition, object tracking, and other computer vision tasks. We have a strong emphasis on combining traditional image processing algorithms with advanced neural network architecture design and deployment. We are constantly looking for ways to expand the application space of image processing algorithms through continued collaboration with other groups and researchers in different fields and even in different universities. By utilizing cutting-edge advancement in artificial intelligence and principled algorithm design, we strive to create solutions for real world problems in a wide variety of domains - healthcare, education and defense to name a few. Academically, our lab emphasizes on researching neural network architecture design and applied mathematical techniques (linear algebra, probabilistic frameworks)for the purpose of image and video analysis.
Active Research
AI for Activity Recognition
Artificial Intelligence for Human Activity Recognition in Classrooms
In collaboration with the SCHOOL OF EDUCATION AND HUMAN DEVELOPMENT, the AIAI Project involves analyses of classroom instructional videos utilizing state of the art neural network architectures for assisting in pedagogical performance evaluation.
Bacterial Biofilms
Image Analyses for Bacterial Biofilms
In collaboration with GAHLMANN LAB - Developing effective automatic image analysis techniques for bacterial biofilms.
Focused Ultrasound Surgery: Image Analyses
Research with an intent for clinical adoption
In collaboration with the SHEYBANI LAB and the DEPARTMENT OF NEUROSURGERY, SCHOOL OF MEDICINE (UVA) , we are researching on image analyses algorithms and data fusion mechanisms to combine image features with clinical variables to help assist neurosurgeons performing focussed ultrasound surgeries PIONEERED AT THE UNIVERSITY OF VIRGINIA.
XAI - Explainable AI
Advancing Explainable AI Techniques
We are researching techniques to advance the concept of EXPLAINABLE AI for image and video analyses.