Image segmentation using tensorflow. Used 2-D convolutional layers to reduce the spatial dimension of an image and then recontruct the spatial dimensions by reducing the depth features. Link to dataset: https://s3-ap-southeast-1.amazonaws.com/he-public-data/contests/SigTuple_data.tar
TODO: Try to upsample using "de-convolutional" layers, http://warmspringwinds.github.io/tensorflow/tf-slim/2016/11/22/upsampling-and-image-segmentation-with-tensorflow-and-tf-slim/