# Faster R-CNN > [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://arxiv.org/abs/1506.01497) ## 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.
## Results and Models | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Config | Download | | :-------------: | :-----: | :-----: | :------: | :------------: | :----: | :-----------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | | R-50-C4 | caffe | 1x | - | - | 35.6 | [config](./faster-rcnn_r50-caffe_c4-1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50-caffe-c4_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_c4_1x_coco/faster_rcnn_r50_caffe_c4_1x_coco_20220316_150152.log.json) | | R-50-DC5 | caffe | 1x | - | - | 37.2 | [config](./faster-rcnn_r50-caffe-dc5_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50-caffe-dc5_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_dc5_1x_coco/faster_rcnn_r50_caffe_dc5_1x_coco_20201030_151909.log.json) | | R-50-FPN | caffe | 1x | 3.8 | | 37.8 | [config](./faster-rcnn_r50-caffe_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50-caffe_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_fpn_1x_coco/faster_rcnn_r50_caffe_fpn_1x_coco_20200504_180032.log.json) | | R-50-FPN | pytorch | 1x | 4.0 | 21.4 | 37.4 | [config](./faster-rcnn_r50_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_1x_coco_20200130_204655.log.json) | | R-50-FPN (FP16) | pytorch | 1x | 3.4 | 28.8 | 37.5 | [config](./faster-rcnn_r50_fpn_amp-1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/fp16/faster_rcnn_r50_fpn_fp16_1x_coco/faster_rcnn_r50_fpn_fp16_1x_coco_20200204-d4dc1471.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/fp16/faster_rcnn_r50_fpn_fp16_1x_coco/faster_rcnn_r50_fpn_fp16_1x_coco_20200204_143530.log.json) | | R-50-FPN | pytorch | 2x | - | - | 38.4 | [config](./faster-rcnn_r50_fpn_2x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50_fpn_2x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_2x_coco/faster_rcnn_r50_fpn_2x_coco_20200504_210434.log.json) | | R-101-FPN | caffe | 1x | 5.7 | | 39.8 | [config](./faster-rcnn_r101-caffe_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r101-caffe_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r101_caffe_fpn_1x_coco/faster_rcnn_r101_caffe_fpn_1x_coco_20200504_180057.log.json) | | R-101-FPN | pytorch | 1x | 6.0 | 15.6 | 39.4 | [config](./faster-rcnn_r101_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r101_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r101_fpn_1x_coco/faster_rcnn_r101_fpn_1x_coco_20200130_204655.log.json) | | R-101-FPN | pytorch | 2x | - | - | 39.8 | [config](./faster-rcnn_r101_fpn_2x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r101_fpn_2x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r101_fpn_2x_coco/faster_rcnn_r101_fpn_2x_coco_20200504_210455.log.json) | | X-101-32x4d-FPN | pytorch | 1x | 7.2 | 13.8 | 41.2 | [config](./faster-rcnn_x101-32x4d_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-32x4d_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_32x4d_fpn_1x_coco/faster_rcnn_x101_32x4d_fpn_1x_coco_20200203_000520.log.json) | | X-101-32x4d-FPN | pytorch | 2x | - | - | 41.2 | [config](./faster-rcnn_x101-32x4d_fpn_2x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-32x4d_fpn_2x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_32x4d_fpn_2x_coco/faster_rcnn_x101_32x4d_fpn_2x_coco_20200506_041400.log.json) | | X-101-64x4d-FPN | pytorch | 1x | 10.3 | 9.4 | 42.1 | [config](./faster-rcnn_x101-64x4d_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-64x4d_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_64x4d_fpn_1x_coco/faster_rcnn_x101_64x4d_fpn_1x_coco_20200204_134340.log.json) | | X-101-64x4d-FPN | pytorch | 2x | - | - | 41.6 | [config](./faster-rcnn_x101-64x4d_fpn_2x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-64x4d_fpn_2x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_64x4d_fpn_2x_coco/faster_rcnn_x101_64x4d_fpn_2x_coco_20200512_161033.log.json) | ## Different regression loss We trained with R-50-FPN pytorch style backbone for 1x schedule. | Backbone | Loss type | Mem (GB) | Inf time (fps) | box AP | Config | Download | | :------: | :------------: | :------: | :------------: | :----: | :----------------------------------------------------: | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | | R-50-FPN | L1Loss | 4.0 | 21.4 | 37.4 | [config](./faster-rcnn_r50_fpn_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50_fpn_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_1x_coco_20200130_204655.log.json) | | R-50-FPN | IoULoss | | | 37.9 | [config](./faster-rcnn_r50_fpn_iou_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50_fpn_iou_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_iou_1x_coco/faster_rcnn_r50_fpn_iou_1x_coco_20200506_095954.log.json) | | R-50-FPN | GIoULoss | | | 37.6 | [config](./faster-rcnn_r50_fpn_giou_1x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50_fpn_giou_1x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_giou_1x_coco_20200505_161120.log.json) | | R-50-FPN | BoundedIoULoss | | | 37.4 | [config](./faster-rcnn_r50_fpn_bounded-iou_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_bounded_iou_1x_coco-98ad993b.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_bounded_iou_1x_coco_20200505_160738.log.json) | ## Pre-trained Models We also train some models with longer schedules and multi-scale training. The users could finetune them for downstream tasks. | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Config | Download | | :-----------------------------------------------------------: | :-----: | :-----: | :------: | :------------: | :----: | :--------------------------------------------------: | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | | [R-50-C4](./faster-rcnn_r50-caffe-c4_ms-1x_coco.py) | caffe | 1x | - | | 35.9 | [config](./faster-rcnn_r50-caffe-c4_ms-1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_c4_mstrain_1x_coco/faster_rcnn_r50_caffe_c4_mstrain_1x_coco_20220316_150527-db276fed.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_c4_mstrain_1x_coco/faster_rcnn_r50_caffe_c4_mstrain_1x_coco_20220316_150527.log.json) | | [R-50-DC5](./faster-rcnn_r50-caffe-dc5_ms-1x_coco.py) | caffe | 1x | - | | 37.4 | [config](./faster-rcnn_r50-caffe-dc5_ms-1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_dc5_mstrain_1x_coco/faster_rcnn_r50_caffe_dc5_mstrain_1x_coco_20201028_233851-b33d21b9.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_dc5_mstrain_1x_coco/faster_rcnn_r50_caffe_dc5_mstrain_1x_coco_20201028_233851.log.json) | | [R-50-DC5](./faster-rcnn_r50-caffe-dc5_ms-3x_coco.py) | caffe | 3x | - | | 38.7 | [config](./faster-rcnn_r50-caffe-dc5_ms-3x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_dc5_mstrain_3x_coco/faster_rcnn_r50_caffe_dc5_mstrain_3x_coco_20201028_002107-34a53b2c.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_dc5_mstrain_3x_coco/faster_rcnn_r50_caffe_dc5_mstrain_3x_coco_20201028_002107.log.json) | | [R-50-FPN](./faster-rcnn_r50-caffe_fpn_ms-2x_coco.py) | caffe | 2x | 3.7 | | 39.7 | [config](./faster-rcnn_r50-caffe_fpn_ms-2x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50-caffe_fpn_ms-2x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_fpn_mstrain_2x_coco/faster_rcnn_r50_caffe_fpn_mstrain_2x_coco_20200504_231813.log.json) | | [R-50-FPN](./faster-rcnn_r50-caffe_fpn_ms-3x_coco.py) | caffe | 3x | 3.7 | | 39.9 | [config](./faster-rcnn_r50-caffe_fpn_ms-3x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r50-caffe_fpn_ms-3x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_caffe_fpn_mstrain_3x_coco/faster_rcnn_r50_caffe_fpn_mstrain_3x_coco_20210526_095054.log.json) | | [R-50-FPN](./faster-rcnn_r50_fpn_ms-3x_coco.py) | pytorch | 3x | 3.9 | | 40.3 | [config](./faster-rcnn_r50_fpn_ms-3x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_mstrain_3x_coco/faster_rcnn_r50_fpn_mstrain_3x_coco_20210524_110822-e10bd31c.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_mstrain_3x_coco/faster_rcnn_r50_fpn_mstrain_3x_coco_20210524_110822.log.json) | | [R-101-FPN](./faster-rcnn_r101-caffe_fpn_ms-3x_coco.py) | caffe | 3x | 5.6 | | 42.0 | [config](./faster-rcnn_r101-caffe_fpn_ms-3x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r101-caffe_fpn_ms-3x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r101_caffe_fpn_mstrain_3x_coco/faster_rcnn_r101_caffe_fpn_mstrain_3x_coco_20210526_095742.log.json) | | [R-101-FPN](./faster-rcnn_r101_fpn_ms-3x_coco.py) | pytorch | 3x | 5.8 | | 41.8 | [config](./faster-rcnn_r101_fpn_ms-3x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_r101_fpn_ms-3x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r101_fpn_mstrain_3x_coco/faster_rcnn_r101_fpn_mstrain_3x_coco_20210524_110822.log.json) | | [X-101-32x4d-FPN](./faster-rcnn_x101-32x4d_fpn_ms-3x_coco.py) | pytorch | 3x | 7.0 | | 42.5 | [config](./faster-rcnn_x101-32x4d_fpn_ms-3x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-32x4d_fpn_ms-3x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_32x4d_fpn_mstrain_3x_coco/faster_rcnn_x101_32x4d_fpn_mstrain_3x_coco_20210524_124151.log.json) | | [X-101-32x8d-FPN](./faster-rcnn_x101-32x8d_fpn_ms-3x_coco.py) | pytorch | 3x | 10.1 | | 42.4 | [config](./faster-rcnn_x101-32x8d_fpn_ms-3x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-32x8d_fpn_ms-3x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_32x8d_fpn_mstrain_3x_coco/faster_rcnn_x101_32x8d_fpn_mstrain_3x_coco_20210604_182954.log.json) | | [X-101-64x4d-FPN](./faster-rcnn_x101-64x4d_fpn_ms-3x_coco.py) | pytorch | 3x | 10.0 | | 43.1 | [config](./faster-rcnn_x101-64x4d_fpn_ms-3x_coco.py) | [model](https://download.openxlab.org.cn/models/mmdetection/FasterR-CNN/weight/faster-rcnn_x101-64x4d_fpn_ms-3x_coco) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_x101_64x4d_fpn_mstrain_3x_coco/faster_rcnn_x101_64x4d_fpn_mstrain_3x_coco_20210524_124528.log.json) | We further finetune some pre-trained models on the COCO subsets, which only contain only a few of the 80 categories. | Backbone | Style | Class name | Pre-traind model | Mem (GB) | box AP | Config | Download | | ------------------------------------------------------------------------ | ----- | ------------------ | -------------------------------------------------------------- | -------- | ------ | ---------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | [R-50-FPN](./faster-rcnn_r50-caffe_fpn_ms-1x_coco-person.py) | caffe | person | [R-50-FPN-Caffe-3x](./faster-rcnn_r50-caffe_fpn_ms-3x_coco.py) | 3.7 | 55.8 | [config](./faster-rcnn_r50-caffe_fpn_ms-1x_coco-person.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco-person/faster_rcnn_r50_fpn_1x_coco-person_20201216_175929-d022e227.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco-person/faster_rcnn_r50_fpn_1x_coco-person_20201216_175929.log.json) | | [R-50-FPN](./faster-rcnn_r50-caffe_fpn_ms-1x_coco-person-bicycle-car.py) | caffe | person-bicycle-car | [R-50-FPN-Caffe-3x](./faster-rcnn_r50-caffe_fpn_ms-3x_coco.py) | 3.7 | 44.1 | [config](./faster-rcnn_r50-caffe_fpn_ms-1x_coco-person-bicycle-car.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco-person-bicycle-car/faster_rcnn_r50_fpn_1x_coco-person-bicycle-car_20201216_173117-6eda6d92.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco-person-bicycle-car/faster_rcnn_r50_fpn_1x_coco-person-bicycle-car_20201216_173117.log.json) | ## Torchvision New Receipe (TNR) Torchvision released its high-precision ResNet models. The training details can be found on the [Pytorch website](https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/). Here, we have done grid searches on learning rate and weight decay and found the optimal hyper-parameter on the detection task. | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Config | Download | | :--------------------------------------------------: | :-----: | :-----: | :------: | :------------: | :----: | :------------------------------------------------: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | | [R-50-TNR](./faster-rcnn_r50-tnr-pre_fpn_1x_coco.py) | pytorch | 1x | - | | 40.2 | [config](./faster-rcnn_r50-tnr-pre_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_tnr-pretrain_1x_coco/faster_rcnn_r50_fpn_tnr-pretrain_1x_coco_20220320_085147-efedfda4.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_tnr-pretrain_1x_coco/faster_rcnn_r50_fpn_tnr-pretrain_1x_coco_20220320_085147.log.json) | ## Citation ```latex @article{Ren_2017, title={Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Ren, Shaoqing and He, Kaiming and Girshick, Ross and Sun, Jian}, year={2017}, month={Jun}, } ```