Create a PyTorch DataSet class to load our own data for Mask-RCNN

[d-diaz]

PyTorch's default Mask-RCNN implementation expects input formatted like this:

During training, the model expects both the input tensors, as well as a targets (list of dictionary), containing:

• boxes (FloatTensor[N, 4]): the ground-truth boxes in [x1, y1, x2, y2] format, with values of x between 0 and W and values of y between 0 and H
• labels (Int64Tensor[N]): the class label for each ground-truth box
• masks (UInt8Tensor[N, H, W]): the segmentation binary masks for each instance

We will probably want to implement a DataSet class to make data loading and any transformations straightforward. Here's a tutorial in PyTorch docs on making a DataSet. Here is an example specific to creating a DataSet for Mask-RCNN .