Relational Embedding for Few-Shot Classification

2021/08/22 by Kang, Dahyun, Kwon, Heeseung, Min, Juhong +1 · 6 citations
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2108.09666

Abstract

We propose to address the problem of few-shot classification by meta-learning "what to observe" and "where to attend" in a relational perspective. Our method leverages relational patterns within and between images via self-correlational representation (SCR) and cross-correlational attention (CCA). Within each image, the SCR module transforms a base feature map into a self-correlation tensor and learns to extract structural patterns from the tensor. Between the images, the CCA module computes cross-correlation between two image representations and learns to produce co-attention between them. Our Relational Embedding Network (RENet) combines the two relational modules to learn relational embedding in an end-to-end manner. In experimental evaluation, it achieves consistent improvements over state-of-the-art methods on four widely used few-shot classification benchmarks of miniImageNet, tieredImageNet, CUB-200-2011, and CIFAR-FS.

Cited by

• Attention-guided few-shot learning for metal surface defect classification
• Slot Attention-based Feature Filtering for Few-Shot Learning
• Efficient Masked Attention Transformer for Few-Shot Classification and Segmentation
• Color as the Impetus: Transforming Few-Shot Learner
• MetaLab: Few-Shot Game Changer for Image Recognition
• Shallow Deep Learning Can Still Excel in Fine-Grained Few-Shot Learning

Related

• Learning to Compare: Relation Network for Few-Shot Learning
• PARN: Position-Aware Relation Networks for Few-Shot Learning
• RelVAE: Generative Pretraining for few-shot Visual Relationship Detection
• Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation
• Dual Relation Mining Network for Zero-Shot Learning