Skip to content

jbpeace/E2ETag

Repository files navigation

E2ETag

E2ETag: An End-to-End Trainable Method for Generating and Detecting Fiducial Markers
Link to our paper: https://www.bmvc2020-conference.com/assets/papers/0890.pdf

Description

E2ETag proposes an end-to-end trainable method for designing, detecting, and enabling fiducial markers with deep learning. Our method is made possible by introducing back-propagatable marker augmentation and superimposition into training. The detector used was a modified DeepLabV3+ encoder which predicts marker's localization, projective pose, and class. The images used for superimposition training were from the ImageNet dataset. Results demonstrate that our method outperforms existing fiducial markers in ideal conditions and especially in the presence of motion blur, contrast fluctuations, noise, and off-axis viewing angles.

Trained Markers (30 Classes)







Sample Detection Visualizations



Citation

If you find our work useful in your research, please consider citing:

@article{peace2020e2etag,
  title={E2ETag: An End-to-End Trainable Method for Generating and Detecting Fiducial Markers},
  author={J. Brennan Peace and Eric Psota and Yanfeng Liu and Lance C. Pérez},
  journal={The 31st British Machine Vision Conference (BMVC)},
  year={2020}
}

About

An End-to-End Trainable Method for Generating and Detecting Fiducial Markers

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages