.png){kind=icon}
Return
IMSDO: Deep metric learning with incremental margin and standard deviation optimization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2025.131376.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We introduce two methods to improve the performance of deep metric learning. • Incremental Margin gradually increases the margin of triplet loss during training. • Inspired by the warmup learning rate, Incremental Margin enables progressive learning. • SDO introduces a loss function to control the standard deviation of the feature maps. • Minimizing the standard deviations of feature maps eliminates irrelevant ones.
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W
Organization
Cited Papers
ShareSaveShareSaveShareSaveDeep metric learning for few-shot image classification: A Review of recent developments.png)
PATTERN RECOGNITION
OAAI
PATTERN RECOGNITIONIF7.6
OAAIShareSaveMulti-proxy feature learning for robust fine-grained visual recognitionPATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
ShareSaveC2RL: Convolutional-Contrastive Learning for Reinforcement Learning Based on Self-Pretraining for Strong AugmentationSENSORSOAAI
SENSORSIF3.5
OAAIShareSaveShareSave
View more