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.

작업
optical flow
설치
pip install libreyolo
지원 티어
추론 전용, v1.8.0부터 지원. 예측, 검증, 내보내기만 지원합니다. 학습 기능은 적용되지 않습니다.
업스트림
Princeton Vision & Learning Lab의 WAFT, BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint). 논문, 소스
라이선스
코드 BSD-3-Clause, 가중치 BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint). 상업적 사용

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.

bash
pip install libreyolo

Predict

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.

Python
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 pixels
Video
from 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)
CLI
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=true

Flow 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

SizeUpstream checkpointParametersFile
ta2, DINOv3 ViT-S/16 encoder56.5 MLibreWAFTt-flow-dinov3.pt
sa2, Depth Anything V2 ViT-S56.9 MLibreWAFTs-flow-dav2.pt
ma1, Depth Anything V2 ViT-S64.2 MLibreWAFTm-flow.pt
la2, Twins-SVT-Large encoder136.0 MLibreWAFTl-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.

Python
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 rate
CLI
libreyolo val model=LibreWAFTm-flow.pt data=my_flow.yaml

Export

작업ONNXTorchScriptExecuTorchTensorRTOpenVINOPaddleMNNRKNNAxelerancnnTFLiteCoreMLCore AI
Optical flowOptical flow to ONNX: 지원하지 않음Optical flow to TorchScript: 지원하지 않음Optical flow to ExecuTorch: 지원하지 않음Optical flow to TensorRT: 지원하지 않음Optical flow to OpenVINO: 지원하지 않음Optical flow to Paddle: 지원하지 않음Optical flow to MNN: 지원하지 않음Optical flow to RKNN: 지원하지 않음Optical flow to Axelera: 지원하지 않음Optical flow to ncnn: 지원하지 않음Optical flow to TFLite: 지원하지 않음Optical flow to CoreML: 지원하지 않음Optical flow to Core AI: 지원하지 않음

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.

파일가중치 라이선스
Optical flow
LibreWAFTm-flow.ptbsd-3-clause
LibreWAFTt-flow-dinov3.ptDINOv3 License
LibreWAFTs-flow-dav2.ptbsd-3-clause
LibreWAFTl-flow-twins.ptbsd-3-clause

위의 모든 파일은 현재 LibreYOLO 조직에 있으며 처음 사용할 때 내려받습니다.

Licensing

내려받는 특정 가중치의 Hugging Face 저장소에서 라이선스를 확인하십시오. LibreYOLO 조직의 모든 체크포인트에는 라이선스가 있으며 한 계열 안에서도 항상 같지는 않습니다. 해당 저장소가 신뢰할 수 있는 기준입니다. 아래 요약은 이 페이지를 마지막으로 검증했을 때 적용된 내용을 설명합니다.

관련 라이선스에 관한 설명이며 법률 자문이 아닙니다. 상업적으로 중요한 사안이라면 라이선스를 직접 읽고 별도의 법률 자문을 받으십시오.

원작
WAFT, Princeton Vision & Learning Lab
업스트림 라이선스
BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint)
업스트림 소스
github.com/princeton-vl/WAFT
LibreYOLO 코드
MIT
가중치
BSD-3-Clause (s, m and l checkpoints); DINOv3 License (t checkpoint), huggingface.co/LibreYOLO에 다시 게시됨
해석
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, m and l are BSD-3-Clause. Their encoders are Apache-2.0 (Depth Anything V2 Small for s and m, a Twins-SVT-Large copy derived from timm for l). Commercial and closed-source use is allowed.
  • t contains 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 using t commercially.

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}
}

github.com/princeton-vl/WAFT에 있는 저자의 인용 블록에서 복사했습니다.

LibreYOLO v1.8.0에서 검증되었습니다. 이 페이지의 지원 표, 체크포인트, 벤치마크 수치는 출시된 라이브러리와 공개된 가중치에서 생성되며 수동으로 작성되지 않습니다.