Robust outdoor scene reconstruction by using wearable omnidirectional vision system (パターン認識・メディア理解)
スポンサーリンク
概要
- 論文の詳細を見る
A 3D modeling method for wide outdoor scenes by using a wearable vision system with one omnidirectional sensor is presented in this paper. We propose a multi-baseline stereo vision method that first estimates the pose of camera, and then reconstructs 3D models of scenes. In addition, GPS is employed to optimize the parameters of camera and determined the scale of estimated extrinsic parameters automatically. This paper also reports some experimental results with the proposed method.
- 社団法人電子情報通信学会の論文
- 2008-11-20
著者
-
IWAI Yoshio
Graduate School of Engineering Science, Osaka University
-
Yachida Masahiko
Faculty Of Information Science And Technology Osaka Institute Of Technology
-
Yachida Masahiko
Faculty Of Engineering Science Osaka University
-
Iwai Yoshio
Graduate School Of Engineering Science Osaka University
-
FAN Xihao
Graduate School of Engineering Science, Osaka University
-
Fan Xihao
Graduate School Of Engineering Science Osaka University
関連論文
- Real-Time Object Detection Using Adaptive Background Model and Margined Sign Correlation
- L_063 Multiple-channel Streaming Delivery for HyperOmni Vision System
- Video Synthesis with High Spatio-Temporal Resolution Using Motion Compensation and Spectral Fusion(Video Generation,Machine Vision Applications)
- Robust outdoor scene reconstruction by using wearable omnidirectional vision system (マルチメディア・仮想環境基盤)
- Robust outdoor scene reconstruction by using wearable omnidirectional vision system (パターン認識・メディア理解)
- Robust outdoor scene reconstruction by using wearable omnidirectional vision system (コンピュータビジョンとイメージメディア)
- Calibration Method for Misaligned Catadioptric Camera(Camera Calibration,Machine Vision Applications)
- Guidance of a Mobile Robot with Environmental Map Using Omnidirectional Image Sensor COPIS (Special Issue on Image Processing and Understanding)
- Construction Method of Efficient Database for Learning-Based Video Super-Resolution