Awesome

PyramidBox: A Context-assisted Single Shot Face Detector.##

Description

I train pyramidbox with pytorch and the result approaches the original paper result,the pretrained model can be downloaded in vgg,the final model can be downloaded in Pyramidbox.the AP in WIDER FACE as following:

	Easy MAP	Medium MAP	hard MAP
origin paper	0.960	0.948	0.888
this repo	0.948	0.938	0.880

the AP in AFW,PASCAL,FDDB as following:

AFW	PASCAL	FDDB
99.65	99.02	0.983

the gap is small with origin paper,I train 120k batch_size 4 which is different from paper,which maybe cause the gap,if you have more gpu ,the final result maybe better.

Requirement

pytorch 0.3
opencv
numpy
easydict

Prepare data

download WIDER face dataset
modify data/config.py
python prepare_wider_data.py

Train

python train.py --batch_size 4  
		--lr 5e-4

Evalution

according to yourself dataset path,modify data/config.py

Evaluate on AFW.

python tools/afw_test.py

Evaluate on FDDB

python tools/fddb_test.py

Evaluate on PASCAL face

python tools/pascal_test.py

test on WIDER FACE

python tools/wider_test.py

Demo

you can test yourself image

python demo.py

Result