Data from MovingCables: Moving Cable Segmentation Method and Dataset

Manipulating cluttered cables, hoses or ropes is challenging for both robots and humans. Humans often simplify these perceptually challenging tasks by pulling or pushing tangled cables and observing the resulting motions. We would like to build a similar system -- in accordance with the interactive perception paradigm -- to aid robotic cable manipulation. A cable motion segmentation method that densely labels moving cable image pixels is a key building block of such a system. We present MovingCables, a moving cable dataset, which we hope will motivate the development and evaluation of cable motion (or semantic) segmentation algorithms. The dataset consists of real-world image sequences automatically annotated with ground truth segmentation masks and optical flow.