Hallucination Improves Few-Shot Object Detection
arXiv:2105.01294
Abstract
Learning to detect novel objects from few annotated examples is of great practical importance. A particularly challenging yet common regime occurs when there are extremely limited examples (less than three). One critical factor in improving few-shot detection is to address the lack of variation in training data. We propose to build a better model of variation for novel classes by transferring the shared within-class variation from base classes. To this end, we introduce a hallucinator network that learns to generate additional, useful training examples in the region of interest (RoI) feature space, and incorporate it into a modern object detection model. Our approach yields significant performance improvements on two state-of-the-art few-shot detectors with different proposal generation procedures. In particular, we achieve new state of the art in the extremely-few-shot regime on the challenging COCO benchmark.
References in corpus (10)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- YOLOv4: Optimal Speed and Accuracy of Object Detection
- YOLOv3: An Incremental Improvement
- Prototypical Networks for Few-shot Learning
- Frustratingly Simple Few-Shot Object Detection
- Delta-encoder: an effective sample synthesis method for few-shot object recognition
- Few-Example Object Detection with Model Communication
- One-Shot Object Detection with Co-Attention and Co-Excitation
- LSTD: A Low-Shot Transfer Detector for Object Detection
- Cooperating RPN's Improve Few-Shot Object Detection