Deep View Morphing

Dinghuang Ji, Junghyun Kwon, Max McFarland, Silvio Savarese; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 2155-2163


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.

Related Material

[pdf] [arXiv]
author = {Ji, Dinghuang and Kwon, Junghyun and McFarland, Max and Savarese, Silvio},
title = {Deep View Morphing},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {July},
year = {2017}