This is a PyTorch re-implementation of CPN (Cascaded Pyramid Network), winner of MSCOCO keypoints2017 challenge. The TensorFlow version can be found here, which is implemented by the paper author.
Method | Base Model | Input Size | BBox | AP @0.5:0.95 | AP @0.5 | AP @0.75 | AP medium | AP large |
---|---|---|---|---|---|---|---|---|
CPN | ResNet-50 | 256x192 | Ground Truth | 71.2 | 91.4 | 79.2 | 68.3 | 75.5 |
CPN | ResNet-50 | 256x192 | Detection Result | 69.4 | 88.2 | 77.0 | 66.0 | 75.9 |
CPN | ResNet-50 | 384x288 | Ground Truth | 73.7 | 91.5 | 80.6 | 70.6 | 79.2 |
CPN | ResNet-50 | 384x288 | Detection Result | 72.2 | 89.1 | 78.9 | 68.1 | 79.4 |
Thanks Tiamo666 for testing ResNet-50-384x288CPN
model.
If you have interests in this repo, welcome to test other model configurations together.
- Clone the repository
git clone https://github.com/GengDavid/pytorch-cpn
We'll call the directory that you cloned ROOT_DIR
.
-
Download MSCOCO2017 images and annotations from http://cocodataset.org/#download. And put images and annotation files follow the struture showed in data/README.md
After placing data and annotation files. Please runlabel_transform.py
atROOT_DIR
to transform the annotation fomat. -
Initialize cocoapi
git submodule init
git submodule update
cd cocoapi/PythonAPI
make
It will build cocoapi tools automatically.
- Install requirement This repo require following dependences.
- PyTorch == 0.4.0
- numpy >= 1.7.1
- scipy >= 0.13.2
- python-opencv >= 3.3.1
- tqdm > 4.11.1
- skimage >= 0.13.1
- Training
cd ROOT_DIR/MODEL_DIR/
python3 train.py
For example, to train CPN with input resolution 256x192, just change directory into ROOT_DIR/256.192.model, and run the script.
For more args, see by using
python train.py --help
cd ROOT_DIR/MODEL_DIR/
python3 test.py -t PRE-TRAINED_MODEL_NAME
-t
meas use which pre-trained model to test.
For more args, see by using
python test.py --help
If you want to test a pre-trained model, please place the pre-trained model into ROOT_DIR/MODEL_DIR/checkpoint
directory. Please make sure your have put the corresponding model into the folder.
For example, to run pre-trained CPN model with input resolution 256x192,
python3 test.py -t 'CPN256x192'
This pre-trained model is provided below.
COCO.res50.256x192.CPN (updated!)
COCO.res50.384x288.CPN
The detection results are tranformed from results in tf version of cpn.
detection_minival
Thanks chenyilun95, bearpaw and last-one for sharing their codes, which helps me a lot to build this repo.
Thanks Tiamo666 for testing ResNet-50-384x288CPN
model.
Thanks mkocabas for helping me test other configurations.
If you have any questions or find some mistakes about this re-implementation, please open an issue to let me know.
If you want to know more details about the original implementation, you can check tf version of cpn.
Thanks Tiamo666 to point it out that the refineNet is implemented in a different way from the original paper(this can reach a higher results, but it will cost more memory).
Codes and results have been updated!
[1] Chen, Y., Wang, Z., Peng, Y., Zhang, Z., Yu, G., Sun, J.: Cascaded pyramid network for multi-person pose estimation. CVPR (2018)