Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation

2021/03/22 by Gongjie Zhang, Zhang, Gongjie, Zhipeng Luo +5
Computer Science · Medicine · #Advanced Neural Network Applications #Artificial Intelligence (cs.AI) #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.2103.11731

openalex publication_date 2021/03/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Few-shot object detection has been extensively investigated by incorporating meta-learning into region-based detection frameworks. Despite its success, the said paradigm is constrained by several factors, such as (i) low-quality region proposals for novel classes and (ii) negligence of the inter-class correlation among different classes. Such limitations hinder the generalization of base-class knowledge for the detection of novel-class objects. In this work, we design Meta-DETR, a novel few-shot detection framework that incorporates correlational aggregation for meta-learning into DETR detection frameworks. Meta-DETR works entirely at image level without any region proposals, which circumvents the constraint of inaccurate proposals in prevalent few-shot detection frameworks. Besides, Meta-DETR can simultaneously attend to multiple support classes within a single feed-forward. This unique design allows capturing the inter-class correlation among different classes, which significantly reduces the misclassification of similar classes and enhances knowledge generalization to novel classes. Experiments over multiple few-shot object detection benchmarks show that the proposed Meta-DETR outperforms state-of-the-art methods by large margins. The implementation codes will be released at https://github.com/ZhangGongjie/Meta-DETR.

Citations

• Adam: A Method for Stochastic Optimization
• Understanding Neural Networks Through Deep Visualization
• The Pascal Visual Object Classes (VOC) Challenge
• Matching Networks for One Shot Learning
• Prototypical Networks for Few-shot Learning
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal\n Networks
• Meta-Learning: A Survey
• Decoupled Weight Decay Regularization
• Finding Task-Relevant Features for Few-Shot Learning by Category Traversal
• Few-Shot Image Recognition by Predicting Parameters from Activations
• Dynamic Few-Shot Visual Learning without Forgetting
• Learning Deep Features for Discriminative Localization
• An Analysis of Scale Invariance in Object Detection - SNIP
• TAFE-Net: Task-Aware Feature Embeddings for Low Shot Learning
• LSTD: A Low-Shot Transfer Detector for Object Detection
• Deep Learning for Generic Object Detection: A Survey
• Boosting Few-Shot Learning With Adaptive Margin Loss
• End-to-End Object Detection with Adaptive Clustering Transformer

Related

• Incremental-DETR: Incremental Few-Shot Object Detection via Self-Supervised Learning
• Top-Related Meta-Learning Method for Few-Shot Object Detection
• Recent Few-Shot Object Detection Algorithms: A Survey with Performance Comparison
• Decoupled DETR For Few-shot Object Detection
• Frustratingly Simple Few-Shot Object Detection