Ezflow

<p align="center"> <br> <img src="./docs/assets/logo.png" height="60" width="60"/> <br> </p> <h1 align="center">EzFlow</h1> <h3 align="center">A modular PyTorch library for optical flow estimation using neural networks</h3> <div align='center'>

<!-- [](https://github.com/neu-vig/ezflow/actions/workflows/linting.yml) --> <!-- [](https://github.com/neu-vig/ezflow/actions/workflows/codecov.yml) -->

Documentation | Tutorials

</div>

Installation

From source (recommended)

git clone https://github.com/neu-vig/ezflow
cd ezflow/
python setup.py install

From PyPI

pip install ezflow

---

Models supported

• [x] DICL
• [x] DCVNet (1 checkpoint)
• [x] FlowNetS
• [x] FlowNetC (3 checkpoints)
• [x] PWCNet (3 checkpoints)
• [x] RAFT (3 checkpoints)
• [x] VCN

Datasets supported

• [x] AutoFlow
• [x] FlyingChairs
• [x] HD1K
• [x] KITTI
• [x] Kubric
• [x] MPI Sintel
• [x] SceneFlow Monkaa
• [x] SceneFlow Driving
• [x] SceneFlow FlyingThings3D
• [x] SceneFlow FlyingThings3D subset

---

Results and Pre-trained checkpoints

• DCVNet | model config | paper

| Training Dataset | Training Config | ckpts | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all | |-----------------------------------------|-------------------------------------------------------------------------|----------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------| | FlyingThings3DSubset + Monkaa + Driving | config | download | 1.90 | 3.35 | 4.75 | 23.41% |

• FlowNetC | model config | arXiv

| Training Dataset | Training Config | ckpts | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all | |------------------|-------------------------------------------------------------------------|----------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------| | Chairs | config | download | 3.41 | 4.94 | 14.84 | 54.23% | | Chairs -> Things | config | download | 2.93 | 4.48 | 12.47 | 45.89% | | Kubric | config | download | 3.57 | 3.96 | 12.11 | 36.35% |

• PWC-Net | model config | arXiv

| Training Dataset | Training Config | ckpts | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all | |------------------|---------------------------------------------------------------------|-------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------| | Chairs | config | download | 3.5 | 4.73 | 17.81 | 51.76% | | Chairs -> Things | config | download| 2.06 | 3.43 | 11.04 | 32.68% | | Kubric | config | download | 3.08 | 3.31 | 9.83 | 21.94% |

• RAFT | model config | arXiv

| Training Dataset | Training Config | ckpts | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all | |------------------|-----------------------------------------------------------------|--------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------| | Chairs | config | download | 2.23 | 4.56 | 10.45 | 38.93% | | Chairs -> Things | config | download | 1.66 | 2.75 | 5.01 | 16.87% | | Kubric | config | download | 2.12 | 2.54 | 6.01 | 17.35% |

---

Additional Information

• KITTI dataset has been evaluated with a center crop of size 1224 x 370.
• FlowNetC and PWC-Net uses padding of size 64 for evaluating the KITTI2015 dataset.
• RAFT and DCVNet uses padding of size 8 for evaluating the Sintel and KITTI2015 datasets.

---

References

• RAFT
• DICL-Flow
• PWC-Net
• FlowNetPytorch
• VCN
• detectron2
• CorrelationLayer
• ptflow

<br> <footer> <a target="_blank" href="https://icons8.com/icon/3Nj3FNnz36Id/pixels">Pixels</a> icon by <a target="_blank" href="https://icons8.com">Icons8</a> </footer>