Abstract | ||
---|---|---|
A novel paradigm in SLAM involving user interaction without precise selection using snakes is presented allowing robot and landmark estimation. SLAM and particularly Vision-SLAM rely on automated and algorithm dependent features for robot and landmark estimation. Interactive user input is often not accounted for which could not only provide semantics through object differentiation, giving better map description as well as on-the-fly decision making for a person or robot. General Purpose computing on Graphics Processing Units produce Gradient Vector Flow forces, needed for real-time snake evolution. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1007/978-3-319-22416-9_3 | Lecture Notes in Artificial Intelligence |
Keywords | Field | DocType |
SLAM,Interactive,GVF,GPGPU | Graphics,Monocular slam,Computer vision,General purpose computing,Computer science,Vector flow,Artificial intelligence,General-purpose computing on graphics processing units,Robot,Landmark,Semantics | Conference |
Volume | ISSN | Citations |
9287 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 9 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
erick noe amezquita lucio | 1 | 0 | 0.34 |
Jun Liu | 2 | 215 | 20.63 |
Tony J. Dodd | 3 | 201 | 20.88 |