Gedas Bertasius
PhD, CIS '19 - Assistant Professor, University of North Carolina, Chapel Hill
Gedas Bertasius is an Assistant Professor in the Department of Computer Science at the University of North Carolina, Chapel Hill. His current research focuses on topics of video understanding, first-person vision, and multi-modal deep learning. Prior to joining UNC, he was a postdoctoral researcher at Facebook AI. Before that, he received his Bachelors Degree in Computer Science from Dartmouth College, and a Ph.D. in Computer Science from the University of Pennsylvania. His recent work was nominated for the CVPR 2020 best paper award.
Publications
Egocentric Basketball Motion Planning from a Single First-Person Image
Publisher IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)
Learning discriminative motion features through detection
Publisher arXiv
Object detection in video with spatiotemporal sampling networks
Publisher Lecture Notes in Computer Science (LNCS)
Am i a Baller? Basketball Performance Assessment from First-Person Videos
Publisher IEEE International Conference on Computer Vision (ICCV)
Convolutional random walk networks for semantic image segmentation
Publisher IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)
First-person action-object detection with EgoNet
Publisher Robotics: Science and Systems (RSS)
Local perturb-and-MAP for structured prediction
Publisher International Conference on Artificial Intelligence and Statistics (AISTATS)
Unsupervised Learning of Important Objects from First-Person Videos
Publisher IEEE International Conference on Computer Vision (ICCV)
Using Cross-Model EgoSupervision to Learn Cooperative Basketball Intention
Publisher IEEE International Conference on Computer Vision Workshops (ICCVW)
Publisher Lecture Notes in Computer Science (LNCS)
First person action-object detection with egonet
Publisher arXiv
Semantic Segmentation with Boundary Neural Fields
Publisher IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)
DeepEdge: A multi-scale bifurcated deep network for top-down contour detection
Publisher IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)
Exploiting egocentric object prior for 3d saliency detection
Publisher arXiv
Publisher IEEE International Conference on Computer Vision (ICCV)