Few-Shot Object Detection Based on Self-Knowledge Distillation

In many fields, due to the lack of large-scale training data, the traditional object detection methods cannot complete the actual work well. The main reason is the overfitting problem and lack of the generalization ability. In this work, we propose a general method to alleviate the overfitting problem in the few-shot object detection. Our work extends Faster R-CNN with self-knowledge distillation algorithm and designs the loss function with attention mechanism, which can improve true detection in the foreground. In this way, object detector can learn an approximate mapping relationship from few samples, which makes the network possess a stronger generalization ability when tackling few images. Through numerous comparative experiments, we demonstrate that our method is general and feasible on VOC and COCO benchmarks datasets with different settings. We provide a new idea for solving the problem of few-shot object detection, and produce an excellent performance of recall rate on few-shot object detection.

Authors

• Yang Li Northwestern Polytechnical University
• Yicheng Gong
• Zhuo Zhang Harbin Institute of Technology (Shenzhen) Pengcheng Laboratory

Keywords

• Training
• Object detection
• Task analysis
• Intelligent systems
• Training data
• Knowledge transfer
• Transfer learning
• Few-shot Object Detection
• Loss Function
• Generalization Ability
• Comparative Experiments
• Attention Mechanism
• Recall Rate
• VOC Dataset
• Information Technology
• Softmax
• Training Phase
• Bounding Box
• Electrical Engineering
• Field Of Computer Vision
• Base Classes
• Student Model
• Object Detection Task
• Field Of Object Detection
• Joint Training
• Tianjin University
• Student Network
• Object Detection Framework
• Few-shot Learning
• Methods In Most Cases
• Citation Information
• Bounding Box Coordinates
• Method In This Paper