Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add our weights for raft_small on C+T #5076

Merged
merged 2 commits into from
Dec 9, 2021
Merged

Add our weights for raft_small on C+T #5076

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
3c07750
Add our weights for raft_small on C+T
NicolasHug Dec 9, 2021
d8431fc
Update torchvision/prototype/models/optical_flow/raft.py
NicolasHug Dec 9, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions references/optical_flow/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,10 +48,18 @@ torchrun --nproc_per_node 1 --nnodes 1 train.py --val-dataset sintel --batch-siz
```

This should give an epe of about 1.3822 on the clean pass and 2.7161 on the
final pass of Sintel. Results may vary slightly depending on the batch size and
the number of GPUs. For the most accurate resuts use 1 GPU and `--batch-size 1`:
final pass of Sintel-train. Results may vary slightly depending on the batch
size and the number of GPUs. For the most accurate resuts use 1 GPU and
`--batch-size 1`:

```
Sintel val clean epe: 1.3822 1px: 0.9028 3px: 0.9573 5px: 0.9697 per_image_epe: 1.3822 f1: 4.0248
Sintel val final epe: 2.7161 1px: 0.8528 3px: 0.9204 5px: 0.9392 per_image_epe: 2.7161 f1: 7.5964
```

You can also evaluate on Kitti train:

```
torchrun --nproc_per_node 1 --nnodes 1 train.py --val-dataset kitti --batch-size 1 --dataset-root $dataset_root --model raft_large --pretrained
Kitti val epe: 4.7968 1px: 0.6388 3px: 0.8197 5px: 0.8661 per_image_epe: 4.5118 f1: 16.0679
```
7 changes: 3 additions & 4 deletions torchvision/models/optical_flow/raft.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,8 +22,7 @@

_MODELS_URLS = {
"raft_large": "https://download.pytorch.org/models/raft_large_C_T_V2-1bb1363a.pth",
# TODO: change to V2 once we upload our own weights
"raft_small": "https://download.pytorch.org/models/raft_small_C_T_V1-ad48884c.pth",
"raft_small": "https://download.pytorch.org/models/raft_small_C_T_V2-01064c6d.pth",
}


Expand Down Expand Up @@ -591,7 +590,7 @@ def raft_large(*, pretrained=False, progress=True, **kwargs):
`RAFT: Recurrent All Pairs Field Transforms for Optical Flow <https://arxiv.org/abs/2003.12039>`_.

Args:
pretrained (bool): TODO not implemented yet
pretrained (bool): Whether to use pretrained weights.
progress (bool): If True, displays a progress bar of the download to stderr
kwargs (dict): Parameters that will be passed to the :class:`~torchvision.models.optical_flow.RAFT` class
to override any default.
Expand Down Expand Up @@ -636,7 +635,7 @@ def raft_small(*, pretrained=False, progress=True, **kwargs):
`RAFT: Recurrent All Pairs Field Transforms for Optical Flow <https://arxiv.org/abs/2003.12039>`_.

Args:
pretrained (bool): TODO not implemented yet
pretrained (bool): Whether to use pretrained weights.
progress (bool): If True, displays a progress bar of the download to stderr
kwargs (dict): Parameters that will be passed to the :class:`~torchvision.models.optical_flow.RAFT` class
to override any default.
Expand Down
26 changes: 22 additions & 4 deletions torchvision/prototype/models/optical_flow/raft.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,8 @@ class Raft_Large_Weights(WeightsEnum):
"recipe": "https://github.com/princeton-vl/RAFT",
"sintel_train_cleanpass_epe": 1.4411,
"sintel_train_finalpass_epe": 2.7894,
"kitti_train_per_image_epe": 5.0172,
"kitti_train_f1-all": 17.4506,
},
)

Expand All @@ -46,6 +48,8 @@ class Raft_Large_Weights(WeightsEnum):
"recipe": "https://github.com/pytorch/vision/tree/main/references/optical_flow",
"sintel_train_cleanpass_epe": 1.3822,
"sintel_train_finalpass_epe": 2.7161,
"kitti_train_per_image_epe": 4.5118,
"kitti_train_f1-all": 16.0679,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Out of curiosity is this the standard F1 score?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not really, it's a kitti-specific metric, I saw it in different papers (and it's what they report in the leaderboards)

"f1": ((epe > 3) & (relative_epe > 0.05)).float().mean().item() * 100,

},
)

Expand Down Expand Up @@ -87,10 +91,25 @@ class Raft_Small_Weights(WeightsEnum):
"recipe": "https://github.com/princeton-vl/RAFT",
"sintel_train_cleanpass_epe": 2.1231,
"sintel_train_finalpass_epe": 3.2790,
"kitti_train_per_image_epe": 7.6557,
"kitti_train_f1-all": 25.2801,
},
)
C_T_V2 = Weights(
# Chairs + Things
url="https://download.pytorch.org/models/raft_small_C_T_V2-01064c6d.pth",
transforms=RaftEval,
meta={
**_COMMON_META,
"recipe": "https://github.com/princeton-vl/RAFT",
"sintel_train_cleanpass_epe": 1.9901,
"sintel_train_finalpass_epe": 3.2831,
"kitti_train_per_image_epe": 7.5978,
"kitti_train_f1-all": 25.2369,
},
)

default = C_T_V1 # TODO: Change to V2 once we upload our own weights
default = C_T_V2


@handle_legacy_interface(weights=("pretrained", Raft_Large_Weights.C_T_V2))
Expand Down Expand Up @@ -143,14 +162,13 @@ def raft_large(*, weights: Optional[Raft_Large_Weights] = None, progress=True, *
return model


# TODO: change to V2 once we upload our own weights
@handle_legacy_interface(weights=("pretrained", Raft_Small_Weights.C_T_V1))
@handle_legacy_interface(weights=("pretrained", Raft_Small_Weights.C_T_V2))
def raft_small(*, weights: Optional[Raft_Small_Weights] = None, progress=True, **kwargs):
"""RAFT "small" model from
`RAFT: Recurrent All Pairs Field Transforms for Optical Flow <https://arxiv.org/abs/2003.12039>`_.

Args:
weights(Raft_Small_weights, optinal): TODO not implemented yet
weights(Raft_Small_weights, optional): pretrained weights to use.
progress (bool): If True, displays a progress bar of the download to stderr
kwargs (dict): Parameters that will be passed to the :class:`~torchvision.models.optical_flow.RAFT` class
to override any default.
Expand Down