This talk will overview the state of research in deep convolutional neural networks for computer vision and describe one of our recent work on extending this paradigm from image data to point-cloud data.
Point-cloud data arises in many applications where sensors produce sets of 3D points, e.g. autonomous driving/flying, robotics, surveying, 3D imaging, etc. Traditional convolutional neural network (CNN) designs for image data are difficult to be extended to point clouds due to the irregular sampling of points in the 3D space. We will present our new network architecture, PointConv, which efficiently implements CNNs on point cloud data and demonstrate dramatic improvements over the state-of-the-art on classification and semantic segmentation tasks.
Fuxin Li is an assistant professor in the School of Electrical Engineering and Computer Science at Oregon State University. His main research interests are deep learning, video object segmentation, multi-target tracking, point cloud deep networks, adversarial deep learning and human understanding of deep learning. Before OSU, he held research positions at University of Bonn and Georgia Institute of Technology He obtained a Ph.D. degree from the Institute of Automation, Chinese Academy of Sciences, in 2009. He won an NSF CAREER award, (co-)won the PASCAL VOC semantic segmentation challenges from 2009-2012, and led a team to the 4th place finish in the DAVIS Video Segmentation challenge 2017. He has published more than 50 papers in computer vision, machine learning and natural language processing.