ReCoNet-PyTorch
ReCoNet-PyTorch copied to clipboard
Possible bug in data processing generating incorrect optical flow
https://github.com/safwankdb/ReCoNet-PyTorch/blob/8d6d98a32968d87c5ae26aed0aae0adb3dfa94f6/totaldata.py#L112-L115
Here the shape of flow has been permuted, so it should be
flow[0, :, :] *= float(flow.shape[1])/originalflow.shape[0]
,
flow[1, :, :] *= float(flow.shape[2])/originalflow.shape[1]