Few-Shot Image Recognition by Predicting Parameters from Activations

06/12/2017
by   Siyuan Qiao, et al.
0

In this paper, we are interested in the few-shot learning problem. In particular, we focus on a challenging scenario where the number of categories is large and the number of examples per novel category is very limited, e.g. 1, 2, or 3. Motivated by the close relationship between the parameters and the activations in a neural network associated with the same category, we propose a novel method that can adapt a pre-trained neural network to novel categories by directly predicting the parameters from the activations. Zero training is required in adaptation to novel categories, and fast inference is realized by a single forward pass. We evaluate our method by doing few-shot image recognition on the ImageNet dataset, which achieves the state-of-the-art classification accuracy on novel categories by a significant margin while keeping comparable performance on the large-scale categories. We also test our method on the MiniImageNet dataset and it strongly outperforms the previous state-of-the-art methods.

READ FULL TEXT
research
03/20/2020

Few-Shot Learning with Geometric Constraints

In this article, we consider the problem of few-shot learning for classi...
research
12/10/2019

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

This paper proposes a multi-layer neural network structure for few-shot ...
research
08/18/2020

Dataset Bias in Few-shot Image Recognition

The goal of few-shot image recognition (FSIR) is to identify novel categ...
research
07/14/2020

Top-Related Meta-Learning Method for Few-Shot Detection

Many meta-learning methods which depend on a large amount of data and mo...
research
07/22/2017

Single-Shot Clothing Category Recognition in Free-Configurations with Application to Autonomous Clothes Sorting

This paper proposes a single-shot approach for recognising clothing cate...
research
06/21/2023

Black-Box Prediction of Flaky Test Fix Categories Using Language Models

Flaky tests are problematic because they non-deterministically pass or f...
research
04/23/2018

Memory Matching Networks for One-Shot Image Recognition

In this paper, we introduce the new ideas of augmenting Convolutional Ne...

Please sign up or login with your details

Forgot password? Click here to reset