Generative Adversarial Network Architectures For Image Synthesis Using Capsule Networks
This addresses the challenge of inefficient training in image synthesis for researchers and practitioners, though it appears incremental as it builds on existing GAN and Capsule Network concepts.
The paper tackles the problem of image synthesis by proposing GAN architectures that use Capsule Networks as critics, resulting in faster learning of the data manifold and synthesis of visually accurate images with significantly fewer training samples and epochs compared to CNN-based GANs.
In this paper, we propose Generative Adversarial Network (GAN) architectures that use Capsule Networks for image-synthesis. Based on the principal of positional-equivariance of features, Capsule Network's ability to encode spatial relationships between the features of the image helps it become a more powerful critic in comparison to Convolutional Neural Networks (CNNs) used in current architectures for image synthesis. Our proposed GAN architectures learn the data manifold much faster and therefore, synthesize visually accurate images in significantly lesser number of training samples and training epochs in comparison to GANs and its variants that use CNNs. Apart from analyzing the quantitative results corresponding the images generated by different architectures, we also explore the reasons for the lower coverage and diversity explored by the GAN architectures that use CNN critics.