Learning to Write Anywhere with Spatial Transformer Image-to-Motion Encoder-Decoder Networks | IEEE Conference Publication | IEEE Xplore