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).Note: the training time is ~13 hours on Titian X.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.Run fetch_data/fetch_region_proposals.m to download the pre-computed region proposals.Run fetch_data/fetch_model_ResNet101.m to download an ImageNet-pre-trained ResNet-101 net.Run fetch_data/fetch_model_ResNet50.m to download an ImageNet-pre-trained ResNet-50 net.Run experiments/script_rfcn_demo.m to apply the R-FCN model on demo images.Run fetch_data/fetch_demo_model_ResNet101.m to download a R-FCN model using ResNet-101 net (trained on VOC 07+12 trainval).Run fetch_data/fetch_caffe_mex_windows_vs2013_cuda75.m to download a compiled Caffe mex (for Windows only).If you are using Linux or you want to compile for Windows, please recompile our Caffe branch.If you are using Windows, you may download a compiled mex file by running fetch_data/fetch_caffe_mex_windows_vs2013_cuda75.m.Caffe build for R-FCN (included in this repository, see external/caffe).Main Results | training data | test data | mAP | time/img (K40) | time/img (Titian X) If you find R-FCN useful in your research, please consider citing: = ,
MATLAB 2014A LICENSE CHOMIKUJ LICENSE
R-FCN is released under the MIT License (refer to the LICENSE file for details).
MATLAB 2014A LICENSE CHOMIKUJ CODE
This code has been tested on Windows 7/8 64 bit, Windows Server 2012 R2, and Ubuntu 14.04, with Matlab 2014a. R-FCN was initially described in a NIPS 2016 paper. R-FCN can natually adopt powerful fully convolutional image classifier backbones, such as ResNets, for object detection.
![matlab 2014a license chomikuj matlab 2014a license chomikuj](https://manglares.rocks/wp-content/uploads/2018/07/Manglares_logo3.png)
![matlab 2014a license chomikuj matlab 2014a license chomikuj](http://materiallab.aisencc.com/wp-content/uploads/sites/5/2014/05/11.jpg)
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 is a region-based object detection framework leveraging deep fully-convolutional networks, which is accurate and efficient. R-FCN: Object Detection via Region-based Fully Convolutional Networksīy Jifeng Dai, Yi Li, Kaiming He, Jian SunĪ Python version of R-FCN supporting joint training is available here.