# Deep View Morphing

## TL;DR

Deep View Morphing introduces a novel CNN architecture for view synthesis that produces high-quality middle views from two images, overcoming issues like texture loss and shape distortion common in previous methods.

## Contribution

The paper presents a new CNN architecture combining rectification, dense correspondence, and view synthesis modules for improved view interpolation.

## Key findings

- Outperforms state-of-the-art CNN view synthesis methods
- Produces more detailed and accurate middle views
- Reduces texture loss and shape distortions

## Abstract

Recently, convolutional neural networks (CNN) have been successfully applied to view synthesis problems. However, such CNN-based methods can suffer from lack of texture details, shape distortions, or high computational complexity. In this paper, we propose a novel CNN architecture for view synthesis called "Deep View Morphing" that does not suffer from these issues. To synthesize a middle view of two input images, a rectification network first rectifies the two input images. An encoder-decoder network then generates dense correspondences between the rectified images and blending masks to predict the visibility of pixels of the rectified images in the middle view. A view morphing network finally synthesizes the middle view using the dense correspondences and blending masks. We experimentally show the proposed method significantly outperforms the state-of-the-art CNN-based view synthesis method.