# Learnable Visual Markers

arXiv: https://arxiv.org/abs/1610.09237 · HTML: https://mlsanity.com/paper/1610.09237/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Oct 2016.

## Built on by (most cited first, 5 of 5)

- [Neural Lens Modeling](https://mlsanity.com/paper/2304.04848/) (2023)
- [NeuralMarker: A Framework for Learning General Marker Correspondence](https://mlsanity.com/paper/2209.08896/) (2022)
- [DeepFormableTag: End-to-end Generation and Recognition of Deformable Fiducial Markers](https://mlsanity.com/paper/2206.08026/) (2022)
- [CylinderTag: An Accurate and Flexible Marker for Cylinder-Shape Objects Pose Estimation Based on Projective Invariants](https://mlsanity.com/paper/2310.13320/) (2023)
- [Ninja Codes: Neurally Generated Fiducial Markers for Stealthy 6-DoF Tracking](https://mlsanity.com/paper/2510.18976/) (2025)

## Abstract (arXiv)

We propose a new approach to designing visual markers (analogous to QR-codes, markers for augmented reality, and robotic fiducial tags) based on the advances in deep generative networks. In our approach, the markers are obtained as color images synthesized by a deep network from input bit strings, whereas another deep network is trained to recover the bit strings back from the photos of these markers. The two networks are trained simultaneously in a joint backpropagation process that takes characteristic photometric and geometric distortions associated with marker fabrication and marker scanning into account. Additionally, a stylization loss based on statistics of activations in a pretrained classification network can be inserted into the learning in order to shift the marker appearance towards some texture prototype. In the experiments, we demonstrate that the markers obtained using our approach are capable of retaining bit strings that are long enough to be practical. The ability to automatically adapt markers according to the usage scenario and the desired capacity as well as the ability to combine information encoding with artistic stylization are the unique properties of our approach. As a byproduct, our approach provides an insight on the structure of patterns that are most suitable for recognition by ConvNets and on their ability to distinguish composite patterns.