Tensor-to-Image: Image-to-Image Translation with Vision Transformers

TL;DR

This paper introduces a vision transformer-based model called tensor-to-image for image-to-image translation, demonstrating its ability to generalize across different problems without modifications.

Contribution

The paper presents a novel transformer-based architecture specifically designed for image translation tasks, highlighting its flexibility and generalization capabilities.

Findings

Model effectively performs image-to-image translation tasks.

Self-attention enables the model to generalize across various problems.

The approach requires no modifications for different applications.

Abstract

Transformers gain huge attention since they are first introduced and have a wide range of applications. Transformers start to take over all areas of deep learning and the Vision transformers paper also proved that they can be used for computer vision tasks. In this paper, we utilized a vision transformer-based custom-designed model, tensor-to-image, for the image to image translation. With the help of self-attention, our model was able to generalize and apply to different problems without a single modification.