Cross-View Image Synthesis with Deformable Convolution and Attention Mechanism

TL;DR

This paper introduces a GAN-based approach utilizing deformable convolution and attention mechanisms to improve cross-view image synthesis, effectively handling large view differences and occlusions to produce more realistic images.

Contribution

It proposes a novel GAN framework with deformable convolution and attention modules specifically designed for cross-view image synthesis, enhancing feature extraction and correspondence learning.

Findings

Outperforms state-of-the-art methods on Dayton dataset

Produces more realistic images with better scene understanding

Effective in handling occlusions and view differences

Abstract

Learning to generate natural scenes has always been a daunting task in computer vision. This is even more laborious when generating images with very different views. When the views are very different, the view fields have little overlap or objects are occluded, leading the task very challenging. In this paper, we propose to use Generative Adversarial Networks(GANs) based on a deformable convolution and attention mechanism to solve the problem of cross-view image synthesis (see Fig.1). It is difficult to understand and transform scenes appearance and semantic information from another view, thus we use deformed convolution in the U-net network to improve the network's ability to extract features of objects at different scales. Moreover, to better learn the correspondence between images from different views, we apply an attention mechanism to refine the intermediate feature map thus generating more realistic images. A large number of experiments on different size images on the Dayton dataset[1] show that our model can produce better results than state-of-the-art methods.