| Installation | Documentation | Tutorials |
forked from gluon-cv
edited by Yuerong Chen and Ning Li
- PASCAL VOC 2007 (Train/Test: 2007+2012trainval/2007test, shorter size: 600)
model | #GPUs | bs | lr | epoch | decay | AP50 |
---|---|---|---|---|---|---|
VGG-RPN-baseline | 2 | 2 | 1e-3 | 20 | 14 | 73.0 |
VGG-RPN-cascade | 2 | 2 | 1e-3 | 20 | 14 | 75.3 |
VGG-RPN-baseline(pruned) | 2 | 2 | 5e-4 | 20 | 14 | 73 |
VGG-RPN-cascade(pruned) | 2 | 2 | 5e-4 | 20 | 14 | 75 |
-
Clone the cascade_rcnn_gluon repository, and we'll call the directory that you cloned cascade_rcnn_gluon into CASCADE_ROOT
git clone https://github.com/zzdang/cascade_rcnn_gluon.git
-
Build cascade_rcnn_gluon
cd $CASCADE_ROOT/ # Follow the gluon-cv installation instructions here: # https://gluon-cv.mxnet.io/ python setup.py install
-
Get the training data
# This will download the pascal_voc dataset python scripts/datasets/pascal_voc.py
-
Download the pretrained models on ImageNet. For VGG-Net(we called vgg16_pruned), the FC layers are pruned and 2048 units per FC layer are remained. In addition, the two FC layers are copied three times for Cascade R-CNN training.
# Download pre-trained model(You can download it use dropbox or baiduyun link) -[dropbox link](https://www.dropbox.com/s/tjgcwqgber2tlxh/VGG_16_fc2048_prune.params?dl=0) -[baiduyun link](https://pan.baidu.com/s/1RgG33zy40ssdWHdhPx0-Kg) passwd: b7ev # copy the pre-trained models to $CASCADE_ROOT/models/ cp /PATH/TO/DOWNLOAD/MODEL $CASCADE_ROOT/models/
-
training for Cascade-RCNN
# training for pruned VGG16 python scripts/detection/cascade_rcnn/train_cascade_rcnn.py --network vgg16_pruned
-
Testing Demo
# testing for pruned VGG16 (VOC dataset) python scripts/detection/cascade_rcnn/demo_cascade_rcnn.py --network cascade_rcnn_vgg16_pruned_voc --pretrained /PATH/TO/TRAINED/MODEL
- Train Cascade-RCNN with VGG16 and pruned vgg16 backbone
- Train Cascade-RCNN with pruned VGG16 backbone(2 batch pre gpu)
- Train faster-RCNN and Cascade-RCNN with resnet50 backbone