A Multi-Modal, Discriminative and Spatially Invariant CNN for RGB-D Object Labeling.
Umar AsifMohammed BennamounFerdous Ahmed SohelPublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2018)
Keyphrases
- multi modal
- d objects
- spatially invariant
- spatially varying
- image sequences
- viewpoint
- object recognition
- three dimensional
- gray scale
- multi view
- multiple views
- active learning
- color images
- semi supervised
- shape descriptors
- motion blur
- photometric stereo
- high dimensional
- image segmentation
- image processing
- machine learning
- multiresolution
- image analysis