Libra R-CNN: Towards Balanced Learning for Object Detection

Abstract

Compared with model architectures, the training process, which is also crucial to the success of detectors, has received relatively less attention in object detection. In this work, we carefully revisit the standard training practice of detectors, and find that the detection performance is often limited by the imbalance during the training process, which generally consists in three levels - sample level, feature level, and objective level. To mitigate the adverse effects caused thereby, we propose Libra R-CNN, a simple but effective framework towards balanced learning for object detection. It integrates three novel components: IoU-balanced sampling, balanced feature pyramid, and balanced L1 loss, respectively for reducing the imbalance at sample, feature, and objective level. Benefitted from the overall balanced design, Libra R-CNN significantly improves the detection performance. Without bells and whistles, it achieves 2.5 points and 2.0 points higher Average Precision (AP) than FPN Faster R-CNN and RetinaNet respectively on MSCOCO.

To appear at CVPR 2019

References in corpus (9)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
• Hybrid Task Cascade for Instance Segmentation
• Non-local Neural Networks
• Acquisition of Localization Confidence for Accurate Object Detection
• SNIPER: Efficient Multi-Scale Training
• Deep Feature Pyramid Reconfiguration for Object Detection

Cited by in corpus (16)

• A Survey of Deep Learning-based Object Detection
• Learning Spatial Fusion for Single-Shot Object Detection
• AugFPN: Improving Multi-scale Feature Learning for Object Detection
• PF-Net: Point Fractal Network for 3D Point Cloud Completion
• FoodLogoDet-1500: A Dataset for Large-Scale Food Logo Detection via Multi-Scale Feature Decoupling Network
• Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection
• HRDNet: High-resolution Detection Network for Small Objects
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Deep neural networks approach to microbial colony detection -- a comparative analysis
• Dual ResGCN for Balanced Scene GraphGeneration
• Effective Fusion Factor in FPN for Tiny Object Detection
• Balance Scene Learning Mechanism for Offshore and Inshore Ship Detection in SAR Images
• Query-based Hard-Image Retrieval for Object Detection at Test Time
• DHARI Report to EPIC-Kitchens 2020 Object Detection Challenge
• ProbaNet: Proposal-balanced Network for Object Detection
• Object Detection on Single Monocular Images through Canonical Correlation Analysis