A PyTorch implementation of SSDs (include original ssd, DRFNet, RefineDet)

• Installation
• Datasets
• Train
• Evaluate
• Performance
• Reference

• Install PyTorch-0.4.0 by selecting your environment on the website and running the appropriate command.
• Clone this repository.
  • Note: We currently only support Python 3+.
• Then download the dataset by following the instructions below.
• Compile the nms and install coco tools:

cd SSD_Pytorch
# if you use anaconda3, maybe you need https://github.com/rbgirshick/py-faster-rcnn/issues/706
./make.sh
pip install pycocotools

Note*: Check you GPU architecture support in utils/build.py, line 131. Default is:

'nvcc': ['-arch=sm_52',

To make things easy, we provide a simple VOC dataset loader that inherits torch.utils.data.Dataset making it fully compatible with the torchvision.datasets API.

# specify a directory for dataset to be downloaded into, else default is ~/data/
sh data/scripts/VOC2007.sh # <directory>

# specify a directory for dataset to be downloaded into, else default is ~/data/
sh data/scripts/VOC2012.sh # <directory>

move all images in VOC2007 and VOC2012 into VOCROOT/VOC0712/JPEGImages
move all annotations in VOC2007 and VOC2012 into VOCROOT/VOC0712/JPEGImages/Annotations
rename and merge some txt VOC2007 and VOC2012 ImageSets/Main/*.txt to VOCROOT/VOC0712/JPEGImages/ImageSets/Main/*.txt
the merged txt list as follows:
2012_test.txt, 2007_test.txt, 0712_trainval_test.txt, 2012_trainval.txt, 0712_trainval.txt

Install the MS COCO dataset at /path/to/coco from official website, default is ~/data/COCO. Following the instructions to prepare minival2014 and valminusminival2014 annotations. All label files (.json) should be under the COCO/annotations/ folder. It should have this basic structure

$COCO/
$COCO/cache/
$COCO/annotations/
$COCO/images/
$COCO/images/test2015/
$COCO/images/train2014/
$COCO/images/val2014/

UPDATE: The current COCO dataset has released new train2017 and val2017 sets which are just new splits of the same image sets.

• First download the fc-reduced VGG-16 PyTorch base network weights at: https://s3.amazonaws.com/amdegroot-models/vgg16_reducedfc.pth
• ResNet pre-trained basenet weight file is available at ResNet50, ResNet101, ResNet152
• By default, we assume you have downloaded the file in the SSD_Pytorch/weights/pretrained_models dir:

mkdir weights
cd weights
mkdir pretrained_models

wget https://s3.amazonaws.com/amdegroot-models/vgg16_reducedfc.pth
wget https://download.pytorch.org/models/resnet50-19c8e357.pth
wget https://download.pytorch.org/models/resnet101-5d3b4d8f.pth
wget https://download.pytorch.org/models/resnet152-b121ed2d.pth
mv download_weights pretrained_models

• To train SSD_Pytorch using the train script simply specify the parameters listed in train.py as a flag or manually change them.

python train.py --cfg ./configs/ssd_vgg_voc.yaml

• Note: All training configs are in ssd_vgg_voc.yaml, you can change it by yourself.

• To evaluate a trained network:

python eval.py --cfg ./configs/ssd_vgg_voc.yaml --weights ./eval_weights

• To detect one images

 # you need put some images in ./images
python demo.py --cfg ./configs/ssd_vgg_voc.yaml --images ./images --save_folder ./output

You can specify the parameters listed in the eval.py or demo.py file by flagging them or manually changing them.

we retrained some models, so it's different from the origin paper size = 300

• Wei Liu, et al. "SSD: Single Shot MultiBox Detector." ECCV2016.
• Original Implementation (CAFFE)
• A list of other great SSD ports that were sources of inspiration (especially the Chainer repo):
  • ssd.pytorch, RFBNet Chainer, torchcv )