Deep SIMBAD: Landmark Ranking-based Scene Descriptor for Highly Compressive Self-localization Across Domains.
Tanaka KanjiPublished in: SSCI (2021)
Keyphrases
- d scene
- ranking algorithm
- video sequences
- dynamic scenes
- compressive sensing
- real world
- image registration
- three dimensional
- single image
- action recognition
- ranking functions
- object recognition
- shape descriptors
- multiple images
- discriminative power
- moving objects
- image sequences
- link analysis
- scene understanding
- multiple objects
- scene images
- omni directional
- outdoor scenes
- scene classification
- video data
- feature extraction
- ranked list
- learning to rank
- input image