Towards Contextual Learning in Few-shot Object Classification

TL;DR

This paper introduces a method for few-shot object classification that leverages contextual information in complex scenes, improving performance over traditional approaches that focus on isolated objects.

Contribution

It proposes two plug-and-play modules that enable existing FSL methods to incorporate contextual learning, grounded in educational science principles.

Findings

Contextual learning outperforms isolated object learning in experiments.

Modules effectively weight important context elements during learning.

Approach demonstrates superior results on Visual Genome and Open Images datasets.

Abstract

Few-shot Learning (FSL) aims to classify new concepts from a small number of examples. While there have been an increasing amount of work on few-shot object classification in the last few years, most current approaches are limited to images with only one centered object. On the opposite, humans are able to leverage prior knowledge to quickly learn new concepts, such as semantic relations with contextual elements. Inspired by the concept of contextual learning in educational sciences, we propose to make a step towards adopting this principle in FSL by studying the contribution that context can have in object classification in a low-data regime. To this end, we first propose an approach to perform FSL on images of complex scenes. We develop two plug-and-play modules that can be incorporated into existing FSL methods to enable them to leverage contextual learning. More specifically, these modules are trained to weight the most important context elements while learning a particular concept, and then use this knowledge to ground visual class representations in context semantics. Extensive experiments on Visual Genome and Open Images show the superiority of contextual learning over learning individual objects in isolation.