Fast View Synthesis with Deep Stereo Vision

TL;DR

This paper introduces a fast and accurate view synthesis method combining stereo vision and CNNs, effectively decomposing the problem into geometry estimation and texture inpainting, outperforming current state-of-the-art methods.

Contribution

It proposes a novel stereo-vision based CNN approach that decomposes view synthesis into geometry and texture tasks, improving speed and accuracy.

Findings

More accurate than existing methods on KITTI dataset

Significantly faster than current state-of-the-art

Effective decomposition of view synthesis into sub-tasks

Abstract

Novel view synthesis is an important problem in computer vision and graphics. Over the years a large number of solutions have been put forward to solve the problem. However, the large-baseline novel view synthesis problem is far from being "solved". Recent works have attempted to use Convolutional Neural Networks (CNNs) to solve view synthesis tasks. Due to the difficulty of learning scene geometry and interpreting camera motion, CNNs are often unable to generate realistic novel views. In this paper, we present a novel view synthesis approach based on stereo-vision and CNNs that decomposes the problem into two sub-tasks: view dependent geometry estimation and texture inpainting. Both tasks are structured prediction problems that could be effectively learned with CNNs. Experiments on the KITTI Odometry dataset show that our approach is more accurate and significantly faster than the current state-of-the-art. The code and supplementary material will be publicly available. Results could be found here https://youtu.be/5pzS9jc-5t0