Graduation 2018 Faculty
Graduation 2018 Group Grad Students
Faculty, student, poster
Undergrad Research Poster 2018
TechConnect 2017
Poster day
Prof Susan Rodger in Classroom
TMLD 2018
Undergrad Research Poster
Undergrad poster presentation
Undergrad research poster
Graduation 2018
Undergrad Poster BreatheMe
Recipients of grad student awards 2017-18
Poster by undergrad Aditya
Undergrad summer research group 2018
Graduation 2018
Graduation 2018
Kristen Stephens Martinez
Undergrad Research Poster 2018

Upcoming Events

Everything You Always Wanted To Know about the NSF ... But Were Afraid to Ask

Duke Computer Science Colloquium
Speaker Name
W. Rance Cleaveland II
Location
LSRC D106
Date and Time
-
The National Science Foundation (NSF) is the premiere federal agency for funding non-medical basic scientific research in the US. The Computing and Communications Foundations (CCF) division, which I head, is part of the Computer and Information Science and Engineering (CISE) directorate of NSF and oversees efforts on the foundations of computing, from algorithms, programing languages, software engineering and hardware design to information theory and future computing paradigms. Each year CCF is responsible for distributing nearly $200m in funding. In this talk I will give a “behind-the-scenes peek” at how CCF, and CISE, operate, and also discuss current and future trends, opportunities and challenges facing the CCF and CISE communities.

Motion Boundaries without Optical Flow

Master's Defense
Speaker Name
Hannah Kim
Location
LSRC D344
Date and Time
-

We propose an architecture based on Convolutional Neural Networks (CNNs) for the detection of motion boundaries from two consecutive images of a video sequence. Existing learning-based approaches start with dense optical flow estimates, which are expensive to compute and often fail near motion boundaries, exactly where they are needed most. In contrast, we explore ways to detect motion boundaries without first computing optical flow. For efficiency, we hypothesize that motion boundaries occur at or near the edges of superpixels in an over-segmentation of the first image.