Disparity interpolation for image synthesis

被引：1

作者：

Dong, YY ^{[1
]}

Wang, H ^{[1
]}

机构：

[1] Nanyang Technol Univ, Sch Elect & Elect Engn, Singapore 639798, Singapore

来源：

PATTERN RECOGNITION LETTERS | 2000年 / 21卷 / 02期

关键词：

image synthesis; rectification; disparity interpolation;

D O I：

10.1016/S0167-8655(99)00164-6

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper reports a novel technique to synthesize new images from two original views snapped by a weakly calibrated perspective camera. The synthesis process is based on disparity interpolation and rectification and it requires neither 3D-scene information nor additional constraints from the two original views. Two algorithms are proposed: direct disparity interpolation (DDI) and projective basis generating algorithm (PBGA). Experiments show that the synthesized wire-frames are physically valid. (C) 2000 Published by Elsevier Science B.V. All rights reserved.

引用

页码：201 / 210

页数：10

共 15 条

[11]

Xu AG, 2018, Epipolar geometry in stereo, motion, and object recognition: a unified approach, DOI [10.1007/978-94-015-8668-9, DOI 10.1007/978]

[12]

ZELLER C, P INT C PATT REC, P132

[13] A ROBUST TECHNIQUE FOR MATCHING 2 UNCALIBRATED IMAGES THROUGH THE RECOVERY OF THE UNKNOWN EPIPOLAR GEOMETRY [J].

ZHANG, ZY ;

DERICHE, R ;

FAUGERAS, O ;

LUONG, QT .

ARTIFICIAL INTELLIGENCE, 1995, 78 (1-2) :87-119

[14]

ZHANG ZY, 1998, P INT C COMP VIS

[15]

ZHANG ZY, 1997, IN PRESS J COMPUT VI

← 1 2 →