Hi, I am a Computer Vision Research Scientist at Amazon.com working on the Amazon Go project, where we are reimagining the grocery shopping experience.
I graduated with a PhD in Computer Science from University of Southern California. I was a graduate research assistant in the Computer Vision Laboratory at Institute for Robotics and Intelligent Systems. My thesis adviser was Prof. Ram Nevatia.
Previously, I earned my M.S. degree in Computer Science from University of Southern California in May, 2011. I also hold a Bachelors in Engineering degree in Electronics and Electrical Communication Engineering and a Masters in Engineering degree in Visual information and Embedded Systems from the Indian Institute of Technology, Kharagpur (better known as IIT Kharagpur). At IIT, I worked in Prof. Somnath Sengupta’s Computer Vision Laboratory.
My research interests are broadly in the field of computer vision, pattern recognition and machine learning. During my PhD, my research focused on recognizing complex human activities in videos, which contain a large number of articulated pose variations in complex scenes. I have developed novel algorithms for combining global statistics of local features, with the high level structure and constraints provided by dynamical graphical models, in a discriminative learning framework. Our algorithms minimize the manual annotation requirements during training by learning probabilistic latent variable models to capture semantic concepts (like human-pose and human-object interactions), and at the same time are flexible enough to incorporate varying styles and duration of activity dynamics. During my PhD, I have been funded by and contributed to the following research programs: DARPA VIRAT, DARPA Minds Eye, DARPA ADAMS and DARPA SMISC.
|—||Prithviraj Banerjee and Ram Nevatia. “Multi-State Discriminative Video Segment Selection for Complex Event Classification“, In Asian Conference on Computer Vision (ACCV), 2014. [Conf. Link] [pdf]|
|—||Prithviraj Banerjee and Ram Nevatia. “Pose Filter based Hidden-CRF models for Activity Detection“, In European Conference on Computer Vision (ECCV), 2014. [Conf. Link] [pdf]|
|—||Prithviraj Banerjee and Ram Nevatia. “Pose Based Activity Recognition using Multiple Kernel Learning“, In IAPR International Conference on Pattern Recognition (ICPR), 2012. [Conf. Link] [pdf]|
|—||Prithviraj Banerjee and Ram Nevatia. “Learning Neighborhood Co-occurrence Statistics of Sparse Features for Human Activity Recognition“, In IEEE International Conference on Advanced Video and Signal-Based Surveillance, 2011. [Conf. Link] [pdf]|
|—||Prithviraj Banerjee and Ram Nevatia. “Dynamics based Trajectory Segmentation for UAV videos“, In IEEE International Conference on Advanced Video and Signal-Based Surveillance, August 2010. [Conf. Link] [pdf]|
|—||Pradeep Natarajan, Prithviraj Banerjee and Ram Nevatia. “Accurate Person Tracking Through Changing Poses for Multi-view Action Recognition“, In ICVGIP, 2009. [pdf]|
|—||Pradeep Natarajan, Prithviraj Banerjee, Furqan M. Khan and Ram Nevatia. “Graphical Framework for Action Recognition using Temporally Dense STIPs“, In IEEE Workshop on Motion and Video Computing, December 2009. [Conf. Link] [pdf]|
|—||Prithviraj Banerjee, Axel Pinz and Somnath Sengupta. “Model generation for robust object tracking based on Temporally Stable Regions“, In IEEE Workshop on Motion and Video Computing, January 2008. [pdf]|
|—||Prithviraj Banerjee and Somanth Sengupta. “Human Motion Detection and Tracking for Video Surveillance“, In National Conference on Communication (NCC), February 2008, IIT Bombay, India. [pdf]|