Skip to content
forked from daijifeng001/R-FCN

R-FCN: Object Detection via Region-based Fully Convolutional Networks

License

Notifications You must be signed in to change notification settings

anguoyang/R-FCN

 
 

Repository files navigation

R-FCN: Object Detection via Region-based Fully Convolutional Networks

By Jifeng Dai, Yi Li, Kaiming He, Jian Sun

Introduction

R-FCN is a region-based object detection framework leveraging deep fully-convolutional networks, which is accurate and efficient. In contrast to previous region-based detectors such as Fast/Faster R-CNN that apply a costly per-region sub-network hundreds of times, our region-based detector is fully convolutional with almost all computation shared on the entire image. R-FCN can natually adopt powerful fully convolutional image classifier backbones, such as ResNets, for object detection.

R-FCN was initially described in an arxiv tech report.

This code has been tested on Windows 7/8 64 bit, Windows Server 2012 R2, and Ubuntu 14.04, with Matlab 2014a.

License

R-FCN is released under the MIT License (refer to the LICENSE file for details).

Citing R-FCN

If you find R-FCN useful in your research, please consider citing:

@article{dai16rfcn,
    Author = {Jifeng Dai, Yi Li, Kaiming He, Jian Sun},
    Title = {{R-FCN}: Object Detection via Region-based Fully Convolutional Networks},
    Journal = {arXiv preprint arXiv:1605.06409},
    Year = {2016}
}

Main Results

               | training data       | test data             | mAP   | time/img (K40) | time/img (Titian X)

-------------------|:-------------------:|:---------------------:|:-----:|:--------------:|:------------------:| R-FCN, ResNet-50 | VOC 07+12 trainval | VOC 07 test | 77.4% | 0.12sec | 0.09sec | R-FCN, ResNet-101 | VOC 07+12 trainval | VOC 07 test | 79.5% | 0.17sec | 0.12sec |

Requirements: software

  1. Caffe build for R-FCN (included in this repository, see external/caffe)
    • If you are using Windows, you may download a compiled mex file by running fetch_data/fetch_caffe_mex_windows_vs2013_cuda75.m
    • If you are using Linux or you want to compile for Windows, please recompile our Caffe branch.
  2. MATLAB 2014a or later

Requirements: hardware

GPU: Titan, Titan X, K40, K80.

Demo

  1. Run fetch_data/fetch_caffe_mex_windows_vs2013_cuda75.m to download a compiled Caffe mex (for Windows only).
  2. Run fetch_data/fetch_demo_model_ResNet101.m to download a R-FCN model using ResNet-101 net (trained on VOC 07+12 trainval).
  3. Run rfcn_build.m.
  4. Run startup.m.
  5. Run experiments/script_rfcn_demo.m to apply the R-FCN model on demo images.

Preparation for Training & Testing

  1. Run fetch_data/fetch_caffe_mex_windows_vs2013_cuda75.m to download a compiled Caffe mex (for Windows only).
  2. Run fetch_data/fetch_model_ResNet50.m to download an ImageNet-pre-trained ResNet-50 net.
  3. Run fetch_data/fetch_model_ResNet101.m to download an ImageNet-pre-trained ResNet-101 net.
  4. Run fetch_data/fetch_region_proposals.m to download the pre-computed region proposals.
  5. Download VOC 2007 and 2012 data to ./datasets.
  6. Run rfcn_build.m.
  7. Run startup.m.

Training & Testing

  1. Run experiments/script_rfcn_VOC0712_ResNet50_OHEM_ss.m to train a model using ResNet-50 net with online hard example mining (OHEM), leveraging selective search proposals. The accuracy should be ~75.4% in mAP.
    • Note: the training time is ~13 hours on Titian X.
  2. Run experiments/script_rfcn_VOC0712_ResNet50_OHEM_rpn.m to train a model using ResNet-50 net with OHEM, leveraging RPN proposals (using ResNet-50 net). The accuracy should be ~77.4% in mAP.
    • Note: the training time is ~13 hours on Titian X.
  3. Run experiments/script_rfcn_VOC0712_ResNet101_OHEM_rpn.m to train a model using ResNet-101 net with OHEM, leveraging RPN proposals (using ResNet-101 net). The accuracy should be ~79.5% in mAP.
    • Note: the training time is ~19 hours on Titian X.
  4. Check other scripts in ./experiments for more settings.

Note:

  • In all the experiments, training is performed on VOC 07+12 trainval, and testing is performed on VOC 07 test.
  • Results are subject to some random variations. We have run 'experiments/script_rfcn_VOC0712_ResNet50_OHEM_rpn.m' for 5 times, the results are 77.1%, 77.3%, 77.7%, 77.9%, and 77.0%. The mean is 77.4%, and the std is 0.39%.
  • Running time is not recorded in the test log (which is slower), but instead in an optimized implementation.

Resources

  1. Experiment logs: DropBox, BaiduYun

If the automatic "fetch_data" fails, you may manually download resouces from:

  1. Pre-complied caffe mex (Windows):
  2. Demo R-FCN model:
  3. ImageNet-pretrained networks:
  4. Pre-computed region proposals:

About

R-FCN: Object Detection via Region-based Fully Convolutional Networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • MATLAB 90.0%
  • C++ 5.8%
  • Cuda 4.2%