Variational Prototyping-Encoder: One-Shot Learning With Prototypical Images

Abstract

In daily life, graphic symbols, such as traffic signs and brand logos, are ubiquitously utilized around us due to its intuitive expression beyond language boundary. We tackle an open-set graphic symbol recognition problem by one-shot classification with prototypical images as a single training example for each novel class. We take an approach to learn a generalizable embedding space for novel tasks. We propose a new approach called variational prototyping-encoder (VPE) that learns the image translation task from real-world input images to their corresponding prototypical images as a meta-task. As a result, VPE learns image similarity as well as prototypical concepts which differs from widely used metric learning based approaches. Our experiments with diverse datasets demonstrate that the proposed VPE performs favorably against competing metric learning based one-shot methods. Also, our qualitative analyses show that our meta-task induces an effective embedding space suitable for unseen data representation.

Effect of Probabilistic Similarity Measure on Metric-Based Few-Shot Classification

“…Since the prototypical network [9] has shown better performance than more complicated few-shot learning models [6][7][8], there have been a number of extensions based on the same structure shown in Figure 1 [10,[13][14][15]. Sung et al [10] added the relation module after the embedding module for more fine-grained classification.…”

“…Based on the prototypical networks, Wertheimer et al [14] use a concatenation of foreground and background vector representations as feature vectors. Kim et al [15] introduced the variational autoencoder (VAE) structure [31] into the embedding module for training the prototype images.…”

“…The 1-shot classification accuracies on GTSRB. The values marked with * are quoted from the VPE paper [15], and the ones with ** are measured by experiments using codes provided by the authors [9].…”

“…The embedding module extracts appropriate features through mapping given input to an embedding space, and the classification module tries to classify newly given samples (query data) with only a few training samples (support data) by using the features from the embedding module. Since a new learning strategy, called episodic learning [8], was proposed to obtain a good embedding function under the few-shot scenario in [8], most of the subsequent works have mainly focused on designing a good embedding function model that can provide an efficient and general representation for recognizing unseen test classes [13][14][15][16][17].…”

Knowledge Graph Transfer Network for Few-Shot Recognition

“…These works often adopt learning-to-learn paradigm that distills knowledge learned from training categories to help learn novel concepts. For example, (Vinyals et al 2016;Snell, Swersky, and Zemel 2017;Sung et al 2018;Kim et al 2019) learn an embedding and metric function from the base categories to well recognize samples in the novel categories. Most of these works evaluate their algorithms on some small-scale datasets, e.g., miniImageNet with 64 base categories, 16 validation categories, 20 novel categories.…”