Abstract

Object skeleton is a useful cue for object detection, complementary to the object contour, as it provides a structural representation to describe the relationship among objectparts. While object skeleton extraction in natural imagesis a very challenging problem, as it requires the extractorto be able to capture both local and global image contex-t to determine the intrinsic scale of each skeleton pixel.Existing methods rely on per-pixel based multi-scale feature computation, which results in difficult modeling and high time consumption. In this paper, we present a fullyconvolutional network with multiple scale-associated side outputs to address this problem. By observing the relationship between the receptive field sizes of the sequential stages in the network and the skeleton scales they can capture, we introduce a scale-associated side output to each stage. We impose supervision to different stages by guiding the scale-associated side outputs toward groundtruth skeletons of different scales. The responses of the multiple scaleassociated side outputs are then fused in a scale-specific way to localize skeleton pixels with multiple scales effectively. Our method achieves promising results on two skeleton extraction datasets, and significantly outperforms other competitors.