WAFT
WAFT, Warping-Alone Field Transforms, is an optical flow method by Wang and Deng (ICLR 2026). It replaces RAFT's cost volume with high-resolution feature warping and a ViT refinement loop. LibreYOLO ports the upstream code and mirrors its four zero-shot checkpoints, one per feature encoder.
- Zadania
- optical flow
- Instalacja
pip install libreyolo- Poziom obsługi
- Tylko inferencja, od wersji v1.8.0. Tylko predykcja, walidacja i eksport. Funkcje trenowania nie mają zastosowania.
- Projekt źródłowy
- WAFT, autorzy: Princeton Vision & Learning Lab, licencja: BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint). Publikacja, kod źródłowy
- Licencje
- Kod: BSD-3-Clause, wagi: BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint). Użycie komercyjne
WAFT drops the correlation volume that RAFT and its descendants
build, and instead warps high-resolution features of the second frame with the
current flow estimate before each refinement step. All four checkpoints are
upstream's zero-shot models (TartanAir, then Chairs, then Things). Pick m,
the a1 checkpoint the upstream README recommends for downstream use, unless
you have a reason to choose a specific encoder.
Install
WAFT needs no optional extra. Everything it imports is in the base install.
pip install libreyoloPredict
Weights download from the LibreYOLO Hugging Face organization on first use.
The class LibreWAFT is also exported from libreyolo if you prefer to build the
model directly.
from libreyolo import LibreYOLO model = LibreYOLO("LibreWAFTm-flow.pt") # weights download on first useresults = model(["frame_0001.jpg", "frame_0002.jpg"]) flow = results[0].flowprint(flow.data.shape) # (H, W, 2) float32, (dx, dy) in pixelsprint(flow.mean_magnitude) # mean displacement in pixelsfrom libreyolo import LibreYOLO model = LibreYOLO("LibreWAFTm-flow.pt") # N frames give N - 1 results; result i is the flow from frame i to i + 1.for result in model("video.mp4", stream=True, save=True, save_flow=True): print(result.flow.max_magnitude)libreyolo predict model=LibreWAFTm-flow.pt source=frame_0001.jpg,frame_0002.jpg save=truelibreyolo predict model=LibreWAFTm-flow.pt source=video.mp4 save=true arrows=true save_flow=trueFlow needs at least two frames: a single image raises. A list of images, a
directory sorted by name, a video or a stream is one sequence, and N frames
give N - 1 results. result.flow is a FlowField with an (H, W, 2) float32
array on the original canvas of the first frame; flow.data[y, x] is
(dx, dy) in pixels, +x right and +y down. WAFT fills no confidence map, so flow.confidence is None. result.boxes is
None. save=True writes the Middlebury colour wheel, arrows=True or
warp=True pick the other views, and save_flow=True also writes each pair's
raw field as a .flo file. The optical flow task
page describes the payload and views in full.
Variants
| Size | Upstream checkpoint | Parameters | File |
|---|---|---|---|
t | a2, DINOv3 ViT-S/16 encoder | 56.5 M | LibreWAFTt-flow-dinov3.pt |
s | a2, Depth Anything V2 ViT-S | 56.9 M | LibreWAFTs-flow-dav2.pt |
m | a1, Depth Anything V2 ViT-S | 64.2 M | LibreWAFTm-flow.pt |
l | a2, Twins-SVT-Large encoder | 136.0 M | LibreWAFTl-flow-twins.pt |
The encoder is the weight variant in the file name, and each size exists only
under the name in the table; any other combination raises with the published
name. LibreWAFT defaults to size m. Prediction runs at native resolution:
the network pads the pair to a multiple of 112, 32 or 16 depending on the
encoder and crops the flow back. Five refinement iterations are baked in, the
upstream evaluation default. imgsz caps the long side (default 8192, in
practice no cap).
Training is not offered: train() raises NotImplementedError.
Validate
val() runs the shared flow validator on a flow dataset (frames per sequence
folder, Middlebury .flo, KITTI 16-bit PNG or .npy ground truth; see the
dataset format). It reports the mean
end-point error metrics/epe, metrics/px1 and metrics/px3 (share of valid
pixels under 1 and 3 px of error) and the KITTI outlier rate
metrics/fl_all.
from libreyolo import LibreYOLO model = LibreYOLO("LibreWAFTm-flow.pt")metrics = model.val(data="my_flow.yaml") print(metrics["metrics/epe"]) # mean end-point error, pixelsprint(metrics["metrics/px1"]) # share of pixels under 1 px errorprint(metrics["metrics/fl_all"]) # KITTI outlier ratelibreyolo val model=LibreWAFTm-flow.pt data=my_flow.yamlExport
| Zadanie | ONNX | TorchScript | ExecuTorch | TensorRT | OpenVINO | Paddle | MNN | RKNN | Axelera | ncnn | TFLite | CoreML | Core AI |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Optical flow | Optical flow to ONNX: brak obsługi | Optical flow to TorchScript: brak obsługi | Optical flow to ExecuTorch: brak obsługi | Optical flow to TensorRT: brak obsługi | Optical flow to OpenVINO: brak obsługi | Optical flow to Paddle: brak obsługi | Optical flow to MNN: brak obsługi | Optical flow to RKNN: brak obsługi | Optical flow to Axelera: brak obsługi | Optical flow to ncnn: brak obsługi | Optical flow to TFLite: brak obsługi | Optical flow to CoreML: brak obsługi | Optical flow to Core AI: brak obsługi |
Export is not available yet for any format. Optical flow export is blocked
until the two-input graph (image1, image2) and the backend vector-rescale
contract of the flow task are implemented for a family.
Checkpoints
Every published weight file for this family.
| Plik | Licencja wag |
|---|---|
| Optical flow | |
| LibreWAFTm-flow.pt | bsd-3-clause |
| LibreWAFTt-flow-dinov3.pt | DINOv3 License |
| LibreWAFTs-flow-dav2.pt | bsd-3-clause |
| LibreWAFTl-flow-twins.pt | bsd-3-clause |
Każdy z powyższych plików jest obecnie dostępny w organizacji LibreYOLO i pobierany przy pierwszym użyciu.
Licensing
Sprawdź licencję w repozytorium konkretnych pobieranych wag na Hugging Face. Każdy checkpoint w organizacji LibreYOLO ma licencję, która nie zawsze jest taka sama w całej rodzinie. To repozytorium jest źródłem rozstrzygającym, a poniższe podsumowanie opisuje stan z chwili ostatniej weryfikacji tej strony.
To opis obowiązujących licencji, a nie porada prawna. Jeśli odpowiedź ma znaczenie komercyjne, przeczytaj licencje i zasięgnij własnej porady prawnej.
- Oryginalna praca
- WAFT, Princeton Vision & Learning Lab
- Licencja projektu źródłowego
- BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint)
- Kod źródłowy projektu
- github.com/princeton-vl/WAFT
- Kod LibreYOLO
- MIT
- Wagi
- BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint), ponownie opublikowane w huggingface.co/LibreYOLO
- Interpretacja
- The license depends on the size, because each checkpoint embeds a different encoder. The s, m and l checkpoints are BSD-3-Clause, a permissive license that allows commercial and closed-source use as long as the copyright notice, conditions and disclaimer are kept on redistribution; their encoders are Apache-2.0 (Depth Anything V2 Small for s and m, Twins-SVT-Large from timm for l), which is also permissive. The t checkpoint contains a DINOv3 ViT-S/16 fine-tuned by the WAFT authors, so Meta's DINOv3 License applies to it: a custom license, not OSI-approved, with its own use restrictions and conditions; read it before relying on t commercially. Upstream publishes no separate weight license, so LibreYOLO hosts the checkpoints under the repository's BSD-3-Clause license plus the encoder license. The code from princeton-vl/WAFT at commit b152ff1 is vendored under libreyolo/models/waft/_vendor/ (BSD-3-Clause); the encoders reuse code already in LibreYOLO (Depth Anything V2, Apache-2.0; DINOv3 from the DEIMv2 vendor, DINOv3 License) and a plain-torch Twins copy derived from timm (Apache-2.0). LibreYOLO's wrapper code is MIT. Weights come from the upstream Google Drive model zoo and are mirrored on the LibreYOLO Hugging Face org without gating. Inference only.
WAFT is the work of Wang and Deng at the Princeton Vision & Learning Lab. The upstream code is vendored under its BSD-3-Clause licence. Upstream publishes no separate weight licence, so LibreYOLO mirrors the checkpoints under the repository's BSD-3-Clause licence plus the licence of the encoder each one embeds, without gating.
s,mandlare BSD-3-Clause. Their encoders are Apache-2.0 (Depth Anything V2 Small forsandm, a Twins-SVT-Large copy derived from timm forl). Commercial and closed-source use is allowed.tcontains a DINOv3 ViT-S/16 fine-tuned by the WAFT authors, so Meta's DINOv3 License applies to it: a custom licence, not OSI-approved, with its own use restrictions and trade-control terms. Read it before usingtcommercially.
Citation
@article{wang2025waft,
title={WAFT: Warping-Alone Field Transforms for Optical Flow},
author={Wang, Yihan and Deng, Jia},
journal={arXiv preprint arXiv:2506.21526},
year={2025}
}Skopiowano z bloku cytowania autorów w github.com/princeton-vl/WAFT.