HOSENet: Higher-Order Semantic Enhancement for Few-Shot Object Detection.
Lingli ZhangKe-Jia ChenXiaomeng ZhouPublished in: PRCV (2) (2020)
Keyphrases
- image processing
- multiscale
- higher order
- object detection
- natural images
- computer vision
- image enhancement
- markov random field
- high level
- semantic knowledge
- pairwise
- semantic web
- face detection
- semantic information
- object categories
- low level features
- scene understanding
- video data
- shot boundary detection
- background subtraction
- sports video
- scene recognition
- detection algorithm
- natural language
- domain ontology
- high order
- visual features
- semantic similarity
- multi class
- pedestrian detection
- low level
- moving objects
- video sequences
- low order
- multimedia
- deformable part models
- lambda calculus