This is an experimental Tensorflow implementation of Faster RCNN - a convnet for object detection with a region proposal network. For details about R-CNN please refer to the paper Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks by Shaoqing Ren, Kaiming He, Ross Girshick, Jian Sun.
-
Requirements for Tensorflow (see: Tensorflow) - Tested r.011
-
Python packages needed to run this repo:
cython
,python-opencv
,easydict
- For training the end-to-end version of Faster R-CNN with VGG16, 3G of GPU memory is sufficient (using CUDNN)
- Clone the Faster R-CNN repository
# Make sure to clone with --recursive
git clone --recursive https://github.com/bigsnarfdude/Faster-RCNN_TF.git
- Build the Cython modules
cd Faster-RCNN_TF/lib make
After successfully completing basic installation, you'll be ready to run the demo.
You are going to need to download a pretrained model that was trained on PASCAL VOC 2007 dataset. Download from either of these locations and put into the Faster-RCNN_TF directory [Google Drive] [Dropbox]
Get pretrained model
wget https://www.dropbox.com/s/cfz3blmtmwj6bdh/VGGnet_fast_rcnn_iter_70000.ckpt?dl=0#
mv VGGnet_fast_rcnn_iter_70000.ckpt?dl=0 VGGnet_fast_rcnn_iter_70000.ckpt
To run the demo
cd Faster-RCNN_TF/tools
python demo.py --model /home/ubuntu/Faster-RCNN_TF/VGGnet_fast_rcnn_iter_70000.ckpt --gpu 0
To run the demo IPython Notebook:
The demo.py will load the above pretrained Faster-RCNN_TF network that includes (VGG_ImageNet base and PASCAL 2007 training) and perform detection on each image in the demo.py list. demo.py is located in the tools folder from the project root. The gpu flag will target the first GPU on your box. You can add images into the data folder and run the demo.py against any of your custom photos once you refactor the image list. https://github.com/bigsnarfdude/Faster-RCNN_TF/blob/master/tools/demo.py
-
Download the training, validation, test data and VOCdevkit
wget https://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tar wget https://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tar wget https://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCdevkit_08-Jun-2007.tar
-
Extract all of these tars into one directory named
VOCdevkit
tar xvf VOCtrainval_06-Nov-2007.tar tar xvf VOCtest_06-Nov-2007.tar tar xvf VOCdevkit_08-Jun-2007.tar
-
It should have this basic structure
$VOCdevkit/ # development kit $VOCdevkit/VOCcode/ # VOC utility code $VOCdevkit/VOC2007 # image sets, annotations, etc. # ... and several other directories ...
-
Create symlinks for the PASCAL VOC dataset
cd $FRCN_ROOT/data ln -s $VOCdevkit VOCdevkit2007
-
Download pre-trained ImageNet models
Download the pre-trained ImageNet models [Google Drive] [Dropbox]
mv VGG_imagenet.npy $FRCN_ROOT/data/pretrain_model/VGG_imagenet.npy
-
Run script to train and test model
cd $FRCN_ROOT ./experiments/scripts/faster_rcnn_end2end.sh GPU_ID VGG16 pascal_voc ./experiments/scripts/faster_rcnn_end2end.sh $DEVICE $DEVICE_ID VGG16 pascal_voc
Classes | AP |
---|---|
aeroplane | 0.698 |
bicycle | 0.788 |
bird | 0.657 |
boat | 0.565 |
bottle | 0.478 |
bus | 0.762 |
car | 0.797 |
cat | 0.793 |
chair | 0.479 |
cow | 0.724 |
diningtable | 0.648 |
dog | 0.803 |
horse | 0.797 |
motorbike | 0.732 |
person | 0.770 |
pottedplant | 0.384 |
sheep | 0.664 |
sofa | 0.650 |
train | 0.766 |
tvmonitor | 0.666 |
mAP | 0.681 |
###References Faster R-CNN caffe version