CAFFE for YOLO-9000

Reference

YOLO9000: Better, Faster, Stronger

https://arxiv.org/abs/1612.08242

https://pjreddie.com/darknet/yolo/

https://github.com/yeahkun/caffe-yolo

Usage

Data preparation

  ln -s /your/path/to/VOCdevkit/ .
  python data/yolo/get_list.py
  cd data/yolo
  # change related path in script convert.sh
  ./convert.sh 

Train

  cd examples/yolo
  # change related path in script train.sh and darknet.prototxt
  ./train.sh

Test

weight

https://drive.google.com/open?id=0Bx1yEotM0mQeM3ZPRmhFSHAydUE

binaryproto

https://drive.google.com/open?id=0Bx1yEotM0mQeTE9XYWJjazZTLUE

some image

http://imgur.com/a/mXhat

  cd build/example/yolo_9000_test/
  ./netforward ../../../examples/yolo/darknet_deploy.prototxt path/to/voc_model_iter_30000.caffemode path/to/voc_train.binaryproto test_list > output.txt

test list is a txt file llike

  /VOCdevkit/VOC2007/JPEGImages/000001.jpg
  /VOCdevkit/VOC2007/JPEGImages/000002.jpg
  /VOCdevkit/VOC2007/JPEGImages/000003.jpg
  ...

it will output <label,x,y,w,h> per image

TODO

world tree reorg layer Image Net Pretrain