Learnable Visual Markers

by   Oleg Grinchuk, et al.

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.


page 4

page 5

page 6

page 7


DeepFaceAR: Deep Face Recognition and Displaying Personal Information via Augmented Reality

Biometric recognition is a popular topic in machine vision. Deep Neural ...

Inverting Visual Representations with Convolutional Networks

Feature representations, both hand-designed and learned ones, are often ...

The zero-rate threshold for adversarial bit-deletions is less than 1/2

We prove that there exists an absolute constant δ>0 such any binary code...

MOBS (Matrices Over Bit Strings) public key exchange

We use matrices over bit strings as platforms for Diffie-Hellman-like pu...

Cantor-solus and Cantor-multus Distributions

The Cantor distribution is obtained from bitstrings; the Cantor-solus di...

Uncertainty Surrogates for Deep Learning

In this paper we introduce a novel way of estimating prediction uncertai...

Please sign up or login with your details

Forgot password? Click here to reset