DeepSFM: Robust Deep Iterative Refinement for Structure From Motion.
Xinlin RenXingkui WeiZhuwen LiYanwei FuYinda ZhangXiangyang XuePublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2024)
Keyphrases
- structure from motion
- iterative refinement
- bundle adjustment
- camera pose estimation
- image sequences
- camera parameters
- point cloud
- missing data
- computer vision
- feature tracking
- scene structure
- relative pose
- camera motion
- reconstruction from multiple images
- focal length
- feature correspondences
- reprojection error
- image based localization
- multibody
- line correspondences
- motion segmentation
- photometric information
- optical flow