Quaternion Generative Adversarial Networks for Inscription Detection in Byzantine Monuments

Quaternion Quantum Neural Network for Classification

“…see for example [127], [129], [97], [82], [132], [20], [122], [74], [32], [116], [55], [24]. Another method, proposed by Gaudet and Maida [19], is to use a residual block:…”

“…A more sophisticated type of architecture is the generative models. Sfikas et al [116] propose a Quaternion Generative Adversarial Network for text detection of inscriptions found on byzantine monuments [118], [117]. The generator is a and 7f), where the activation function of the last layer sums the output of real and imaginary parts of the quaternion to produce a real-valued output.…”

“…https://github.com/Orkis-Research/Pytorch-Quaternion-Neural-Networks Sfikas et al [116] GAN https://github.com/sfikas/quaternion-gan A partial approach in this direction is to parametrize the hypercomplex convolution [70], [71]. Let W be a convolution kernel, of size k × k, x a s-dimensional input, and y a ddimensional output, then:…”

Hypercomplex Image-to-Image Translation

“…Indeed, state-of-the-art generative models usually comprise tens of million parameters and are often employed with multidimensional inputs such as color images or multichannel audio signals [17], [31]. The quaternion-valued variational autoencoder and the family of quaternion generative adversarial networks have demonstrated to obtain comparable performance while reducing the storage memory amount due to the parameters reduction [24]- [26], [32]. Encouraged by these results, we propose to expolit novel PHNNs methods to define a more advanced generative model for image-to-image translation.…”

Keyword spotting in handwritten document images using supervised and unsupervised representations

“…• Chapter 5 includes an application of adversarial features represented by quaternion descriptors for text spotting in the wild [44]. In this respect, KWS is not explicitly addressed as the main problem.…”