FlowControl: Optical Flow Based Visual Servoing
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020
Abstract: One-shot imitation is the vision of robot programming from a single demonstration, rather than by tedious construction of computer code. We present a practical method for realizing one-shot imitation for manipulation tasks, exploiting modern learning-based optical flow to perform real-time visual servoing. Our approach, which we call FlowControl, continuously tracks a demonstration video, using a specified foreground mask to attend to an object of interest. Using RGB-D observations, FlowControl requires no 3D object models, and is easy to set up. FlowControl inherits great robustness to visual appearance from decades of work in optical flow. We exhibit FlowControl on a range of problems, including ones requiring very precise motions, and ones requiring the ability to generalize.
Paper
Images and movies
See also
BibTex reference
@InProceedings{AB20, author = "M. Argus and L. Hermann and J. Long and T. Brox", title = "FlowControl: Optical Flow Based Visual Servoing", booktitle = "IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)", month = " ", year = "2020", url = "http://lmbweb.informatik.uni-freiburg.de/Publications/2020/AB20" }