Authors:
Tewodros Habtegebrial
1
;
Kiran Varanasi
2
;
Christian Bailer
2
and
Didier Stricker
3
Affiliations:
1
Department of Computer Science, TU Kaiserslautern, Erwin-Schrdinger-Str. 1, Kaiserslautern and Germany
;
2
German Research Center for Artificial Intelligence, Trippstadter Str. 122, Kaiserslautern and Germany
;
3
Department of Computer Science, TU Kaiserslautern, Erwin-Schrdinger-Str. 1, Kaiserslautern, Germany, German Research Center for Artificial Intelligence, Trippstadter Str. 122, Kaiserslautern and Germany
Keyword(s):
Novel View Synthesis, Convolutional Neural Networks, Stereo Vision.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Computer Vision, Visualization and Computer Graphics
;
Geometry and Modeling
;
Image-Based Modeling
;
Motion, Tracking and Stereo Vision
;
Pattern Recognition
;
Software Engineering
;
Stereo Vision and Structure from Motion
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.