A Unified Framework for Attention-Based Few-Shot Object Detection

TL;DR

This paper introduces a flexible, unified framework for attention-based Few-Shot Object Detection, enabling consistent comparison of different attention mechanisms within various detection architectures.

Contribution

It provides a comprehensive, adaptable framework that standardizes evaluation of attention mechanisms in FSOD, facilitating clearer performance comparisons.

Findings

Reimplemented multiple attention mechanisms within the framework.

Compared various attention techniques under consistent conditions.

Highlighted the impact of different attention mechanisms on FSOD performance.

Abstract

Few-Shot Object Detection (FSOD) is a rapidly growing field in computer vision. It consists in finding all occurrences of a given set of classes with only a few annotated examples for each class. Numerous methods have been proposed to address this challenge and most of them are based on attention mechanisms. However, the great variety of classic object detection frameworks and training strategies makes performance comparison between methods difficult. In particular, for attention-based FSOD methods, it is laborious to compare the impact of the different attention mechanisms on performance. This paper aims at filling this shortcoming. To do so, a flexible framework is proposed to allow the implementation of most of the attention techniques available in the literature. To properly introduce such a framework, a detailed review of the existing FSOD methods is firstly provided. Some different attention mechanisms are then reimplemented within the framework and compared with all other parameters fixed.