A Comparative Review of Recent Few-Shot Object Detection Algorithms
arXiv:2111.00201
Abstract
Few-shot object detection, learning to adapt to the novel classes with a few labeled data, is an imperative and long-lasting problem due to the inherent long-tail distribution of real-world data and the urgent demands to cut costs of data collection and annotation. Recently, some studies have explored how to use implicit cues in extra datasets without target-domain supervision to help few-shot detectors refine robust task notions. This survey provides a comprehensive overview from current classic and latest achievements for few-shot object detection to future research expectations from manifold perspectives. In particular, we first propose a data-based taxonomy of the training data and the form of corresponding supervision which are accessed during the training stage. Following this taxonomy, we present a significant review of the formal definition, main challenges, benchmark datasets, evaluation metrics, and learning strategies. In addition, we present a detailed investigation of how to interplay the object detection methods to develop this issue systematically. Finally, we conclude with the current status of few-shot object detection, along with potential research directions for this field.
References in corpus (17)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- YOLOv4: Optimal Speed and Accuracy of Object Detection
- Recurrent Neural Network Regularization
- Microsoft COCO Captions: Data Collection and Evaluation Server
- Temporal Ensembling for Semi-Supervised Learning
- An Overview of Deep Semi-Supervised Learning
- Augmentation for small object detection
- Meta-learning for Few-shot Natural Language Processing: A Survey
- Resolving Class Imbalance in Object Detection with Weighted Cross Entropy Losses
- Restoring Negative Information in Few-Shot Object Detection
- One-Shot Object Detection without Fine-Tuning
- Cooperating RPN's Improve Few-Shot Object Detection
- Weakly-supervised Learning of Mid-level Features for Pedestrian Attribute Recognition and Localization
- MM-FSOD: Meta and metric integrated few-shot object detection
- A Concise Review of Recent Few-shot Meta-learning Methods
- Review. Machine learning techniques for traffic sign detection
- AFD-Net: Adaptive Fully-Dual Network for Few-Shot Object Detection