Few-Shot Object Detection: A Comprehensive Survey

TL;DR

This survey comprehensively reviews the state of few-shot object detection, categorizing approaches, discussing performance improvements, and analyzing benchmarks to highlight current challenges and promising trends in the field.

Contribution

It provides a detailed categorization of few-shot object detection methods, summarizes key concepts for improving performance, and analyzes benchmark results and evaluation challenges.

Findings

Identification of common challenges in evaluation protocols

Summary of promising trends in few-shot object detection

Analysis of benchmark results across different approaches

Abstract

Humans are able to learn to recognize new objects even from a few examples. In contrast, training deep-learning-based object detectors requires huge amounts of annotated data. To avoid the need to acquire and annotate these huge amounts of data, few-shot object detection aims to learn from few object instances of new categories in the target domain. In this survey, we provide an overview of the state of the art in few-shot object detection. We categorize approaches according to their training scheme and architectural layout. For each type of approaches, we describe the general realization as well as concepts to improve the performance on novel categories. Whenever appropriate, we give short takeaways regarding these concepts in order to highlight the best ideas. Eventually, we introduce commonly used datasets and their evaluation protocols and analyze reported benchmark results. As a result, we emphasize common challenges in evaluation and identify the most promising current trends in this emerging field of few-shot object detection.