raft_large

torchvision.models.optical_flow.raft_large(*, pretrained=False, progress=True, **kwargs)

RAFT model from RAFT: Recurrent All Pairs Field Transforms for Optical Flow.

Please see the example below for a tutorial on how to use this model.

Parameters

• pretrained (bool) –

  Whether to use weights that have been pre-trained on FlyingChairs + FlyingThings3D with two fine-tuning steps:

  • one on Sintel + FlyingThings3D

  • one on KittiFlow.

  This corresponds to the C+T+S/K strategy in the paper.

• progress (bool) – If True, displays a progress bar of the download to stderr.

Returns

The model.

Return type

nn.Module

Examples using raft_large:

Optical Flow: Predicting movement with the RAFT model