How Shift Equivariance Impacts Metric Learning for Instance Segmentation.
Josef Lorenz RumbergerXiaoyan YuPeter HirschMelanie DohmenVanessa Emanuela GuarinoAshkan MokarianLisa MaisJan FunkeDagmar KainmuellerPublished in: ICCV (2021)
Keyphrases
- metric learning
- distance metric
- distance metric learning
- machine learning and pattern recognition
- semi supervised
- dimensionality reduction
- learning tasks
- pairwise
- semi supervised clustering
- image segmentation
- distance function
- multi task
- fully supervised
- mahalanobis metric
- segmentation algorithm
- feature space
- low dimensional
- semi definite programming
- multi class
- prior knowledge