TR2017-114
Fusion of multi-angular aerial Images based on epipolar geometry and matrix completion
-
- "Fusion of multi-angular aerial Images based on epipolar geometry and matrix completion", IEEE International Conference on Image Processing (ICIP), DOI: 10.1109/ICIP.2017.8296471, September 2017.BibTeX TR2017-114 PDF
- @inproceedings{Ma2017sep,
- author = {Ma, Yanting and Liu, Dehong and Mansour, Hassan and Kamilov, Ulugbek and Taguchi, Yuichi and Boufounos, Petros T. and Vetro, Anthony},
- title = {Fusion of multi-angular aerial Images based on epipolar geometry and matrix completion},
- booktitle = {IEEE International Conference on Image Processing (ICIP)},
- year = 2017,
- month = sep,
- doi = {10.1109/ICIP.2017.8296471},
- url = {https://www.merl.com/publications/TR2017-114}
- }
,
- "Fusion of multi-angular aerial Images based on epipolar geometry and matrix completion", IEEE International Conference on Image Processing (ICIP), DOI: 10.1109/ICIP.2017.8296471, September 2017.
-
MERL Contacts:
-
Research Area:
Digital Video
Abstract:
We consider the problem of fusing multiple cloud-contaminated aerial images of a 3D scene to generate a cloud-free image, where the images are captured from multiple unknown view angles. In order to fuse these images, we propose an end-to-end framework incorporating epipolar geometry and low-rank matrix completion. In particular, we first warp the multi-angular images to single-angle ones based on the estimated fundamental matrices that relate the multi-angular images according to their projective relations to the 3D scene. Then we formulate the fusion process of the warpped images as a low-rank matrix completion problem where each column of the matrix corresponds to a vectorized image with missing entries corresponding to cloud or occluded areas. Results using DigitalGlobe high spatial resolution images demonstrate that our algorithm outperforms existing approaches.
Related News & Events
-
NEWS MERL presents 5 papers at ICIP 2017, Anthony Vetro serves as general co-chair Date: September 17, 2017 - September 20, 2017
Where: Beijing, China
MERL Contacts: Petros T. Boufounos; Dehong Liu; Hassan Mansour; Huifang Sun; Anthony Vetro
Research Areas: Computer Vision, Computational Sensing, Digital VideoBrief- MERL presented 5 papers at the IEEE International Conference on Image Processing (ICIP), which was held in Beijing, China from September 17-20, 2017. ICIP is a flagship conference of the IEEE Signal Processing Society and approximately 1300 people attended the event. Anthony Vetro served as General Co-chair for the conference.