# Training Region-based Object Detectors with Online Hard Example Mining

arXiv: https://arxiv.org/abs/1604.03540 · HTML: https://mlsanity.com/paper/1604.03540/

## Key facts

- Cited by 579 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Apr 2016.

## Built on by (most cited first, 30 of 579)

- [AdaFace: Quality Adaptive Margin for Face Recognition](https://mlsanity.com/paper/2204.00964/) (2022)
- [Occ3D: A Large-Scale 3D Occupancy Prediction Benchmark for Autonomous Driving](https://mlsanity.com/paper/2304.14365/) (2023)
- [End-to-End Semi-Supervised Object Detection with Soft Teacher](https://mlsanity.com/paper/2106.09018/) (2021)
- [Unbiased Teacher for Semi-Supervised Object Detection](https://mlsanity.com/paper/2102.09480/) (2021)
- [DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation](https://mlsanity.com/paper/2111.14887/) (2021)
- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [VideoLLM-online: Online Video Large Language Model for Streaming Video](https://mlsanity.com/paper/2406.11816/) (2024)
- [Localizing Visual Sounds the Hard Way](https://mlsanity.com/paper/2104.02691/) (2021)
- [FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding](https://mlsanity.com/paper/2103.05950/) (2021)
- [EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies](https://mlsanity.com/paper/2303.14535/) (2023)
- [Delving into Localization Errors for Monocular 3D Object Detection](https://mlsanity.com/paper/2103.16237/) (2021)
- [General Instance Distillation for Object Detection](https://mlsanity.com/paper/2103.02340/) (2021)
- [LaneRCNN: Distributed Representations for Graph-Centric Motion Forecasting](https://mlsanity.com/paper/2101.06653/) (2021)
- [Improving the Training of Rectified Flows](https://mlsanity.com/paper/2405.20320/) (2024)
- [Semi-supervised Left Atrium Segmentation with Mutual Consistency Training](https://mlsanity.com/paper/2103.02911/) (2021)
- [A Survey of Modern Deep Learning based Object Detection Models](https://mlsanity.com/paper/2104.11892/) (2021)
- [Improving Generalization in Federated Learning by Seeking Flat Minima](https://mlsanity.com/paper/2203.11834/) (2022)
- [Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes](https://mlsanity.com/paper/2101.06085/) (2021)
- [PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers](https://mlsanity.com/paper/2206.02066/) (2022)
- [Fourier Contour Embedding for Arbitrary-Shaped Text Detection](https://mlsanity.com/paper/2104.10442/) (2021)
- [TransVOD: End-to-End Video Object Detection with Spatial-Temporal Transformers](https://mlsanity.com/paper/2201.05047/) (2022)
- [FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation](https://mlsanity.com/paper/2102.12867/) (2021)
- [ArtiBoost: Boosting Articulated 3D Hand-Object Pose Estimation via Online Exploration and Synthesis](https://mlsanity.com/paper/2109.05488/) (2021)
- [A Survey on Long-Tailed Visual Recognition](https://mlsanity.com/paper/2205.13775/) (2022)
- [A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization](https://mlsanity.com/paper/2407.09359/) (2024)
- [Context Decoupling Augmentation for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2103.01795/) (2021)
- [Learning Fast Sample Re-weighting Without Reward Data](https://mlsanity.com/paper/2109.03216/) (2021)
- [R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis](https://mlsanity.com/paper/2203.17261/) (2022)
- [Unbiased Teacher v2: Semi-supervised Object Detection for Anchor-free and Anchor-based Detectors](https://mlsanity.com/paper/2206.09500/) (2022)
- [Ground-aware Monocular 3D Object Detection for Autonomous Driving](https://mlsanity.com/paper/2102.00690/) (2021)

## Abstract (arXiv)

The field of object detection has made significant advances riding on the wave of region-based ConvNets, but their training procedure still includes many heuristics and hyperparameters that are costly to tune. We present a simple yet surprisingly effective online hard example mining (OHEM) algorithm for training region-based ConvNet detectors. Our motivation is the same as it has always been -- detection datasets contain an overwhelming number of easy examples and a small number of hard examples. Automatic selection of these hard examples can make training more effective and efficient. OHEM is a simple and intuitive algorithm that eliminates several heuristics and hyperparameters in common use. But more importantly, it yields consistent and significant boosts in detection performance on benchmarks like PASCAL VOC 2007 and 2012. Its effectiveness increases as datasets become larger and more difficult, as demonstrated by the results on the MS COCO dataset. Moreover, combined with complementary advances in the field, OHEM leads to state-of-the-art results of 78.9% and 76.3% mAP on PASCAL VOC 2007 and 2012 respectively.