Can Scale-Consistent Monocular Depth Be Learned in a Self-Supervised Scale-Invariant Manner?
Lijun WangYifan WangLinzhao WangYunlong ZhanYing WangHuchuan LuPublished in: ICCV (2021)
Keyphrases
- scale invariant
- scale invariance
- scale space
- depth cues
- natural images
- keypoints
- affine invariant
- multiscale
- image structure
- image sequences
- interest point detectors
- local binary pattern
- texture images
- feature transformation
- depth information
- scale selection
- single scale
- invariant features
- pose estimation
- scale space theory
- translation and scale invariant
- scale invariant features
- affine invariance
- feature selection
- object recognition
- optical flow
- scale spaces
- depth map
- color information
- texture classification