FlowSeek

FlowSeek is an optical flow network by Poggi and Tosi (ICCV 2025) that adds a frozen Depth Anything V2 encoder and depth-derived motion bases to a SEA-RAFT style recurrent network. LibreYOLO ports it with four sizes; t and s are Apache-2.0, m and l are for non-commercial use only.

Задачі
optical flow
Встановлення
pip install libreyolo
Рівень підтримки
Лише інференс, починаючи з v1.8.0. Лише передбачення, валідація та експорт. Функції навчання не застосовуються.
Першоджерело
FlowSeek, автори: University of Bologna, ліцензія Apache-2.0 (t and s checkpoints); CC-BY-NC-4.0 (m and l checkpoints). Стаття, джерело
Ліцензії
Код: Apache-2.0, ваги: Apache-2.0 (t and s checkpoints); CC-BY-NC-4.0 (m and l checkpoints). Комерційне використання

FlowSeek borrows geometry knowledge from a depth foundation model: the Depth Anything V2 encoder features and a set of motion bases computed from depth feed the recurrent flow update. The hosted checkpoints are upstream's final generalist fine-tune (TartanAir, Chairs, Things, then a mixed Sintel, KITTI, HD1K and Things stage), the one upstream evaluates on Spring and LayeredFlow. Pick t or s for commercial work; m and l carry a non-commercial encoder.

Install

FlowSeek 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 LibreFlowSeek is also exported from libreyolo if you prefer to build the model directly.

Python
from libreyolo import LibreYOLO model = LibreYOLO("LibreFlowSeekt-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("LibreFlowSeekt-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=LibreFlowSeekt-flow.pt source=frame_0001.jpg,frame_0002.jpg save=truelibreyolo predict model=LibreFlowSeekt-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. FlowSeek 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

Upstream trains two networks; s and l reuse the t and m weights with more refinement iterations, as upstream's evaluation configs do.

SizeContext netDepth Anything V2 encoderIterationsWeights licence
tResNet-18ViT-S4Apache-2.0
sResNet-18ViT-S12Apache-2.0
mResNet-34ViT-B4CC BY-NC 4.0, non-commercial
lResNet-34ViT-B12CC BY-NC 4.0, non-commercial

Files are LibreFlowSeek<size>-flow.pt. LibreFlowSeek defaults to size m, so pass a t or s checkpoint explicitly if you need commercial-use weights. Inference runs at native resolution, padded to a multiple of 8 inside the network, and each side of the padded canvas must be at least 128 px. imgsz caps the long side.

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("LibreFlowSeekt-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=LibreFlowSeekt-flow.pt data=my_flow.yaml

Export

ЗадачаONNXTorchScriptExecuTorchTensorRTOpenVINOPaddleMNNRKNNAxelerancnnTFLiteCoreMLCore AI
Optical flowOptical flow to ONNX: не підтримуєтьсяOptical flow to TorchScript: не підтримуєтьсяOptical flow to ExecuTorch: не підтримуєтьсяOptical flow to TensorRT: не підтримуєтьсяOptical flow to OpenVINO: не підтримуєтьсяOptical flow to Paddle: не підтримуєтьсяOptical flow to MNN: не підтримуєтьсяOptical flow to RKNN: не підтримуєтьсяOptical flow to Axelera: не підтримуєтьсяOptical flow to ncnn: не підтримуєтьсяOptical flow to TFLite: не підтримуєтьсяOptical flow to CoreML: не підтримуєтьсяOptical flow to Core AI: не підтримується

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.

ФайлЛіцензія ваг
Optical flow
LibreFlowSeekt-flow.ptapache-2.0
LibreFlowSeeks-flow.ptapache-2.0
LibreFlowSeekm-flow.ptcc-by-nc-4.0
LibreFlowSeekl-flow.ptcc-by-nc-4.0

Кожен наведений вище файл уже доступний у організації LibreYOLO і завантажується під час першого використання.

Licensing

Перевіряйте ліцензію в репозиторії Hugging Face конкретних ваг, які завантажуєте. Кожна контрольна точка в організації LibreYOLO має ліцензію, і вона не завжди однакова для всього сімейства. Цей репозиторій є авторитетним джерелом, а наведене нижче резюме описує умови на момент останньої перевірки сторінки.

Це опис відповідних ліцензій, а не юридична консультація. Якщо відповідь має комерційне значення, самостійно прочитайте ліцензії та зверніться по юридичну консультацію.

Оригінальна робота
FlowSeek, University of Bologna
Ліцензія першоджерела
Apache-2.0 (t and s checkpoints); CC-BY-NC-4.0 (m and l checkpoints)
Джерело першоджерела
github.com/mattpoggi/flowseek
Код LibreYOLO
MIT
Ваги
Apache-2.0 (t and s checkpoints); CC-BY-NC-4.0 (m and l checkpoints), повторно опубліковано на huggingface.co/LibreYOLO
Тлумачення
The two checkpoint licenses are not interchangeable. Each FlowSeek checkpoint embeds a Depth Anything V2 encoder. The t and s checkpoints (one network, s runs 12 refinement iterations instead of 4) embed the ViT-S encoder and are Apache-2.0: they can be used in commercial and closed-source products, with the license text and attribution notices kept on redistribution. The m and l checkpoints embed the ViT-B encoder, which is CC BY-NC 4.0, so they are non-commercial and need attribution; treat them as research weights. Upstream publishes no separate checkpoint license, so the repository's Apache-2.0 license is applied as an implied grant, limited by the encoder license. The code at commit ba640737 is Apache-2.0 with BSD-3-Clause parts from SEA-RAFT, vendored under libreyolo/models/flowseek/_vendor/; LibreYOLO's wrapper code is MIT. Weights come from the upstream get_weights.sh links and are mirrored on the LibreYOLO Hugging Face org, the m and l cards with a non-commercial banner, without gating. Training is out of scope.

FlowSeek is the work of Poggi and Tosi at the University of Bologna. Its code is Apache-2.0 with BSD-3-Clause parts from SEA-RAFT, vendored in LibreYOLO.

The checkpoint licence depends on the size, because each checkpoint embeds a Depth Anything V2 encoder:

  • t and s embed the ViT-S encoder and are Apache-2.0: commercial and closed-source use is allowed.
  • m and l embed the ViT-B encoder, which is CC BY-NC 4.0: for non-commercial use only, with attribution. Their model cards carry a non-commercial banner and LibreYOLO prints a notice when they load.

Upstream publishes no separate checkpoint licence, so the repository's Apache-2.0 licence is applied as an implied grant, limited by the encoder licence. The weights are mirrored on the LibreYOLO Hugging Face organization without gating.

Citation

@InProceedings{Poggi_2025_ICCV,
    author    = {Poggi, Matteo and Tosi, Fabio},
    title     = {FlowSeek: Optical Flow Made Easier with Depth Foundation Models and Motion Bases},
    booktitle = {Proceedings of the International Conference on Computer Vision (ICCV)},
    year      = {2025},
}

Скопійовано з блоку цитування авторів на сторінці github.com/mattpoggi/flowseek.

Перевірено з LibreYOLO v1.8.0. Таблиці підтримки, контрольні точки й результати бенчмарків на цій сторінці згенеровано з випущеної бібліотеки та опублікованих ваг, а не написано вручну.