GyroFlow+: Gyroscope-Guided Unsupervised Deep Homography and Optical Flow Learning

Haipeng Li, Kunming Luo, Bing Zeng, Shuaicheng Liu

Manuscript

Online Version

Author Version

GHOF Dataset

• For Trainset, the download link is GoogleDrive. Put the data into ./dataset/GHOF_Train, and the contents of directories are as follows:

./dataset/GOF_Train
├── sample_0
│   ├── img1.png
│   ├── img2.png
│   ├── gyro_homo.npy
├── sample_1
│   ├── img1.png
│   ├── img2.png
│   ├── gyro_homo.npy
.....................
├── sample_9900
│   ├── img1.png
│   ├── img2.png
│   ├── gyro_homo.npy

• Benchmark consists of GHOF-Clean and GHOF-Final is available at GoogleDrive
• Download the two BMKs and put to root path

Test Demo

• Download pretrain model step_418600_homo into experiments

python test.py --model_dir experiments/ --restore_file experiments/step_418600_homo.pth

Citation

@inproceedings{li2021gyroflow,
  title={Gyroflow: gyroscope-guided unsupervised optical flow learning},
  author={Li, Haipeng and Luo, Kunming and Liu, Shuaicheng},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages={12869--12878},
  year={2021}
}

@article{li2024gyroflow+,
  title={Gyroflow+: Gyroscope-guided unsupervised deep homography and optical flow learning},
  author={Li, Haipeng and Luo, Kunming and Zeng, Bing and Liu, Shuaicheng},
  journal={International Journal of Computer Vision},
  pages={1--19},
  year={2024},
  publisher={Springer}
}