E2ETag: An End-to-End Trainable Method for Generating and Detecting Fiducial Markers
This work addresses the limitations of fiducial markers for applications in computer vision and augmented reality, offering a more robust solution for real-world environments, though it is incremental as it builds on existing marker detection concepts.
The paper tackles the problem of fiducial marker performance degrading in challenging real-world conditions like motion blur and poor exposure by introducing E2ETag, an end-to-end trainable method for generating and detecting markers, which outperforms existing methods in both ideal and adverse scenarios with significant improvements in detection accuracy.
Existing fiducial markers solutions are designed for efficient detection and decoding, however, their ability to stand out in natural environments is difficult to infer from relatively limited analysis. Furthermore, worsening performance in challenging image capture scenarios - such as poor exposure, motion blur, and off-axis viewing - sheds light on their limitations. E2ETag introduces an end-to-end trainable method for designing fiducial markers and a complimentary detector. By introducing back-propagatable marker augmentation and superimposition into training, the method learns to generate markers that can be detected and classified in challenging real-world environments using a fully convolutional detector network. Results demonstrate that E2ETag outperforms existing methods in ideal conditions and performs much better in the presence of motion blur, contrast fluctuations, noise, and off-axis viewing angles. Source code and trained models are available at https://github.com/jbpeace/E2ETag.