Deep SIMBAD: Active Landmark-based Self-localization Using Ranking -based Scene Descriptor.
Tanaka KanjiPublished in: CoRR (2021)
Keyphrases
- d scene
- single image
- ranking algorithm
- three dimensional
- scene understanding
- web search
- multiple images
- image sequences
- dynamic scenes
- complex scenes
- ranking functions
- discriminative power
- omni directional
- object recognition
- scene classification
- video sequences
- pan tilt zoom camera
- local binary pattern
- shape descriptors
- input image
- multiple objects
- learning to rank
- background subtraction
- image descriptors
- action recognition
- feature extraction