RepMet: Representative-Based Metric Learning for Classification and Few-Shot Object Detection.
Leonid KarlinskyJoseph ShtokSivan HararyEli SchwartzAmit AidesRogério Schmidt FerisRaja GiryesAlexander M. BronsteinPublished in: CVPR (2019)
Keyphrases
- metric learning
- distance metric learning
- object detection
- nearest neighbor classification
- machine learning and pattern recognition
- feature space
- image classification
- distance metric
- margin maximization
- pattern recognition
- distance function
- learning tasks
- pairwise
- supervised learning
- semi supervised
- classification accuracy
- feature extraction
- machine learning
- decision trees
- text classification
- model selection
- unsupervised learning
- video sequences
- learning process
- face recognition
- kernel learning
- multi task
- support vector
- visual features
- data points
- data mining