Image-based Synthesis and Re-Synthesis of Viewpoints Guided by 3D Models

被引：19

作者：

Rematas, Konstantinos ^{[1
]}

Ritschel, Tobias ^{[2
,3
]}

Fritz, Mario ^{[2
]}

Tuytelaars, Tinne ^{[1
]}

机构：

[1] Katholieke Univ Leuven, IMinds, Louvain, Belgium

[2] Max Planck Inst Informat, Munich, Germany

[3] Univ Saarland, D-66123 Saarbrucken, Germany

来源：

2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2014年

关键词：

D O I：

10.1109/CVPR.2014.498

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We propose a technique to use the structural information extracted from a set of 3D models of an object class to improve novel-view synthesis for images showing unknown instances of this class. These novel views can be used to "amplify" training image collections that typically contain only a low number of views or lack certain classes of views entirely (e. g. top views). We extract the correlation of position, normal, reflectance and appearance from computer-generated images of a few exemplars and use this information to infer new appearance for new instances. We show that our approach can improve performance of state-of-the-art detectors using real-world training data. Additional applications include guided versions of inpainting, 2D-to-3D conversion, super-resolution and non-local smoothing.

引用

页码：3898 / 3905

页数：8