A Two-Stage Approach to Few-Shot Learning for Image Recognition

Abstract

This paper proposes a multi-layer neural network structure for few-shot image recognition of novel categories. The proposed multi-layer neural network architecture encodes transferable knowledge extracted from a large annotated dataset of base categories. This architecture is then applied to novel categories containing only a few samples. The transfer of knowledge is carried out at the feature-extraction and the classification levels distributed across the two training stages. In the first-training stage, we introduce the relative feature to capture the structure of the data as well as obtain a low-dimensional discriminative space. Secondly, we account for the variable variance of different categories by using a network to predict the variance of each class. Classification is then performed by computing the Mahalanobis distance to the mean-class representation in contrast to previous approaches that used the Euclidean distance. In the second-training stage, a category-agnostic mapping is learned from the mean-sample representation to its corresponding class-prototype representation. This is because the mean-sample representation may not accurately represent the novel category prototype. Finally, we evaluate the proposed network structure on four standard few-shot image recognition datasets, where our proposed few-shot learning system produces competitive performance compared to previous work. We also extensively studied and analyzed the contribution of each component of our proposed framework.

arXiv commentTo Appear in IEEE Transactions on Image Processing

References in corpus (21)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• A Simple Neural Attentive Meta-Learner
• Generalizing from a Few Examples: A Survey on Few-Shot Learning
• Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning
• On First-Order Meta-Learning Algorithms
• Few-Shot Learning with Graph Neural Networks
• Delta-encoder: an effective sample synthesis method for few-shot object recognition
• Learning to Remember Rare Events
• Learning feed-forward one-shot learners
• Do We Need More Training Data?
• Few-Shot Learning with Metric-Agnostic Conditional Embeddings
• Rapid Adaptation with Conditionally Shifted Neurons
• Generative Adversarial Residual Pairwise Networks for One Shot Learning
• Attentive Recurrent Comparators
• Sample-to-Sample Correspondence for Unsupervised Domain Adaptation
• Small Sample Learning in Big Data Era
• Unsupervised Domain Adaptation using Regularized Hyper-graph Matching

Cited by in corpus (14)

• A Comprehensive Overview and Survey of Recent Advances in Meta-Learning
• Information Symmetry Matters: A Modal-Alternating Propagation Network for Few-Shot Learning
• Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
• Meta-Learned Confidence for Few-shot Learning
• Learning Adaptive Classifiers Synthesis for Generalized Few-Shot Learning
• A Closer Look at Prototype Classifier for Few-shot Image Classification
• FeTrIL: Feature Translation for Exemplar-Free Class-Incremental Learning
• Graph-based Interpolation of Feature Vectors for Accurate Few-Shot Classification
• Dizygotic Conditional Variational AutoEncoder for Multi-Modal and Partial Modality Absent Few-Shot Learning
• Few-Shot Open-Set Recognition using Meta-Learning
• Few-shot Learning for Domain-specific Fine-grained Image Classification
• Learning to Compare Relation: Semantic Alignment for Few-Shot Learning
• NAS-TC: Neural Architecture Search on Temporal Convolutions for Complex Action Recognition
• Task Attended Meta-Learning for Few-Shot Learning