I am currently a research engineer at BlippAR where we tame deep neural networks to augment the reality. Before that I had been working at Artec on face recognition from depth maps. My research interests include (but are not limited to) machine learning, probabilistic graphical models, and their applications to computer vision.
I successfully survived through PhD! I prepared my thesis at Moscow State University under apprenticeship of the dark lord Dmitry Vetrov. It is devoted to structured-output learning for semantic segmentation of images and point clouds under various degree of supervision.
See the comprehensive list of my publications on Google Scholar Citations.
R. Shapovalov, D. Vetrov, A. Osokin, P. Kohli. “Multi-utility Learning: Structured-output Learning with Multiple Annotation-specific Loss Functions”. EMMCVPR. Hong Kong, 2015. [pdf ArXiv:1406.5910]
R. Shapovalov. “Structural learning methods for collective labelling problems ”. [In Russian] PhD Thesis, Lomonosov Moscow State University, 2014. [pdf]
A. Velizhev, R. Shapovalov, K. Schindler. “Implicit shape models for object detection in 3D point clouds”. ISPRS Congress. Melbourne, 2012. [pdf]
R. Shapovalov, A. Velizhev. “Cutting-Plane Training of Non-associative Markov Network for 3D Point Cloud Segmentation”. IEEE 3DIMPVT. Hangzhou, 2011. [project, pdf, slides-pdf, slides-pptx, poster-pdf, data]
I have some code on github.
GML LidarK Library — the library for LIDAR data processing. Currently, only the indexing data structure is implemented. It allows performing spatial queries in 3D space. The code is C++, MATLAB wrapper is also available.
GML BOLT — the toolkit for on-line learning from imbalanced streams. It contains an on-line implementation of the Random Forest algorithm.
Non-research interestsI enjoy snowboarding, cycling, intellectual games and editing Wikipedia.
© Roman Shapovalov, 6-Aug-2016. Powered by X10 Hosting