Parameterized Synthetic Image Data Set for Fisheye Lens
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Standard
Parameterized Synthetic Image Data Set for Fisheye Lens. / Zhen, Chen; Georgiadis, Anthimos.
Proceedings - 2018 5th International Conference on Information Science and Control Engineering, ICISCE 2018. ed. / Y Cheng; S. Li; Y. Dai. IEEE - Institute of Electrical and Electronics Engineers Inc., 2018. p. 370-374 8612582.Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Parameterized Synthetic Image Data Set for Fisheye Lens
AU - Zhen, Chen
AU - Georgiadis, Anthimos
N1 - Conference code: 5
PY - 2018/7/2
Y1 - 2018/7/2
N2 - Based on different projection geometry, a fisheye image can be presented as a parameterized non-rectilinear image. Deep neural networks(DNN) is one of the solutions for extracting the parameters for fisheye image feature expression. However, a number of images are required for training a reasonable prediction model for DNN. In this paper, we propose to extend the scale of the dataset using parameterized synthetic images which effectively boost the diversity of samples and avoid the limitation on the scale. To simulate different viewing angles and distances, we adopt controllable parameterized projection processes on transformation. The reliability of the proposed method is tested with the image captured by a fisheye camera. The synthetic fisheye image dataset is the first dataset that is developed by existing labeled perspective images. It is accessible via: http://www2.leuphana.de/misl/fisheye-data-set/.
AB - Based on different projection geometry, a fisheye image can be presented as a parameterized non-rectilinear image. Deep neural networks(DNN) is one of the solutions for extracting the parameters for fisheye image feature expression. However, a number of images are required for training a reasonable prediction model for DNN. In this paper, we propose to extend the scale of the dataset using parameterized synthetic images which effectively boost the diversity of samples and avoid the limitation on the scale. To simulate different viewing angles and distances, we adopt controllable parameterized projection processes on transformation. The reliability of the proposed method is tested with the image captured by a fisheye camera. The synthetic fisheye image dataset is the first dataset that is developed by existing labeled perspective images. It is accessible via: http://www2.leuphana.de/misl/fisheye-data-set/.
KW - Fisheye lens
KW - Image processing
KW - Neural network
KW - Synthetic data set
KW - Engineering
UR - http://www.scopus.com/inward/record.url?scp=85062108273&partnerID=8YFLogxK
U2 - 10.1109/ICISCE.2018.00084
DO - 10.1109/ICISCE.2018.00084
M3 - Article in conference proceedings
AN - SCOPUS:85062108273
SP - 370
EP - 374
BT - Proceedings - 2018 5th International Conference on Information Science and Control Engineering, ICISCE 2018
A2 - Cheng, Y
A2 - Li, S.
A2 - Dai, Y.
PB - IEEE - Institute of Electrical and Electronics Engineers Inc.
T2 - 5th International Conference on Information Science and Control Engineering - ICISCE 2018
Y2 - 20 July 2018 through 22 July 2018
ER -