A Survey of Deep Learning for Low-Shot Object Detection

Abstract

Object detection is a fundamental task in computer vision and image processing. Current deep learning based object detectors have been highly successful with abundant labeled data. But in real life, it is not guaranteed that each object category has enough labeled samples for training. These large object detectors are easy to overfit when the training data is limited. Therefore, it is necessary to introduce few-shot learning and zeroshot learning into object detection, which can be named low-shot object detection together. Low-Shot Object Detection (LSOD) aims to detect objects from a few or even zero labeled data, which can be categorized into few-shot object detection (FSOD) and zero-shot object detection (ZSD), respectively. This paper conducts a comprehensive survey for deep learning based FSOD and ZSD. First, this survey classifies methods for FSOD and ZSD into different categories and discusses the pros and cons of them. Second, this survey reviews dataset settings and evaluation metrics for FSOD and ZSD, then analyzes the performance of different methods on these benchmarks. Finally, this survey discusses future challenges and promising directions for FSOD and ZSD.

Few-Shot Object Detection: A Comprehensive Survey

“…Although other surveys on FSOD are available [35], [36], [37], [38], [39], [40], they do not cover as many publications related to FSOD as we do. Works [36], [38], [39] are broader surveys, also addressing self-supervised, weakly supervised, and/or zero-shot learning and do not focus as much on FSOD.…”

“…Although other surveys on FSOD are available [35], [36], [37], [38], [39], [40], they do not cover as many publications related to FSOD as we do. Works [36], [38], [39] are broader surveys, also addressing self-supervised, weakly supervised, and/or zero-shot learning and do not focus as much on FSOD. Works [35], [36] only cover earlier work on FSOD and hence are somewhat outdated since at least some of the currently best performing approaches on common benchmarks are missing.…”