Learning to Refine Object Contours with a Top-Down Fully Convolutional Encoder-Decoder Network

Re-assign community