TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Faster R-CNN: Towards Real-Time Object Detection with Regi...

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Shaoqing Ren, Kaiming He, Ross Girshick, Jian Sun

2015-06-04NeurIPS 2015 12Region ProposalReal-Time Object DetectionVessel DetectionRobust Object Detection2D Object DetectionObject Detection
PaperPDFCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCode(official)CodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCode(official)CodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCodeCode

Abstract

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet and Fast R-CNN have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN into a single network by sharing their convolutional features---using the recently popular terminology of neural networks with 'attention' mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while achieving state-of-the-art object detection accuracy on PASCAL VOC 2007, 2012, and MS COCO datasets with only 300 proposals per image. In ILSVRC and COCO 2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been made publicly available.

Results

TaskDatasetMetricValueModel
Object CountingCARPKMAE39.88Faster R-CNN (2015)
Object CountingCARPKRMSE47.67Faster R-CNN (2015)
Object DetectionCOCO-OAverage mAP16.4Faster R-CNN (ResNet-50-FPN)
Object DetectionCOCO-OEffective Robustness-0.41Faster R-CNN (ResNet-50-FPN)
Object DetectionPKU-DDD17-CarmAP5080.2Faster-RCNN
Object DetectionUA-DETRACmAP58.45Faster R-CNN
Object DetectionPASCAL VOC 2007 (15+5)FPS7Faster R-CNN
Object DetectionPASCAL VOC 2007 (15+5)MAP73.2Faster R-CNN
Object DetectionCityscapesmPC [AP]15.4Baseline
3DCOCO-OAverage mAP16.4Faster R-CNN (ResNet-50-FPN)
3DCOCO-OEffective Robustness-0.41Faster R-CNN (ResNet-50-FPN)
3DPKU-DDD17-CarmAP5080.2Faster-RCNN
3DUA-DETRACmAP58.45Faster R-CNN
3DPASCAL VOC 2007 (15+5)FPS7Faster R-CNN
3DPASCAL VOC 2007 (15+5)MAP73.2Faster R-CNN
3DCityscapesmPC [AP]15.4Baseline
2D ClassificationCOCO-OAverage mAP16.4Faster R-CNN (ResNet-50-FPN)
2D ClassificationCOCO-OEffective Robustness-0.41Faster R-CNN (ResNet-50-FPN)
2D ClassificationPKU-DDD17-CarmAP5080.2Faster-RCNN
2D ClassificationUA-DETRACmAP58.45Faster R-CNN
2D ClassificationPASCAL VOC 2007 (15+5)FPS7Faster R-CNN
2D ClassificationPASCAL VOC 2007 (15+5)MAP73.2Faster R-CNN
2D ClassificationCityscapesmPC [AP]15.4Baseline
2D Object DetectionSARDet-100Kbox mAP49F-RCNN
2D Object DetectionCOCO-OAverage mAP16.4Faster R-CNN (ResNet-50-FPN)
2D Object DetectionCOCO-OEffective Robustness-0.41Faster R-CNN (ResNet-50-FPN)
2D Object DetectionPKU-DDD17-CarmAP5080.2Faster-RCNN
2D Object DetectionUA-DETRACmAP58.45Faster R-CNN
2D Object DetectionPASCAL VOC 2007 (15+5)FPS7Faster R-CNN
2D Object DetectionPASCAL VOC 2007 (15+5)MAP73.2Faster R-CNN
2D Object DetectionCityscapesmPC [AP]15.4Baseline
16kCOCO-OAverage mAP16.4Faster R-CNN (ResNet-50-FPN)
16kCOCO-OEffective Robustness-0.41Faster R-CNN (ResNet-50-FPN)
16kPKU-DDD17-CarmAP5080.2Faster-RCNN
16kUA-DETRACmAP58.45Faster R-CNN
16kPASCAL VOC 2007 (15+5)FPS7Faster R-CNN
16kPASCAL VOC 2007 (15+5)MAP73.2Faster R-CNN
16kCityscapesmPC [AP]15.4Baseline

Related Papers

A Real-Time System for Egocentric Hand-Object Interaction Detection in Industrial Domains2025-07-17RS-TinyNet: Stage-wise Feature Fusion Network for Detecting Tiny Objects in Remote Sensing Images2025-07-17Decoupled PROB: Decoupled Query Initialization Tasks and Objectness-Class Learning for Open World Object Detection2025-07-17Dual LiDAR-Based Traffic Movement Count Estimation at a Signalized Intersection: Deployment, Data Collection, and Preliminary Analysis2025-07-17Vision-based Perception for Autonomous Vehicles in Obstacle Avoidance Scenarios2025-07-16Tomato Multi-Angle Multi-Pose Dataset for Fine-Grained Phenotyping2025-07-15ECORE: Energy-Conscious Optimized Routing for Deep Learning Models at the Edge2025-07-08Beyond One Shot, Beyond One Perspective: Cross-View and Long-Horizon Distillation for Better LiDAR Representations2025-07-07