• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    迪恩网络公众号

chuanqi305/MobileNetv2-SSDLite: Caffe implementation of SSD and SSDLite detectio ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

chuanqi305/MobileNetv2-SSDLite

开源软件地址(OpenSource Url):

https://github.com/chuanqi305/MobileNetv2-SSDLite

开源编程语言(OpenSource Language):

Python 80.2%

开源软件介绍(OpenSource Introduction):

MobileNetv2-SSDLite

Caffe implementation of SSD detection on MobileNetv2, converted from tensorflow.

Prerequisites

Tensorflow and Caffe version SSD is properly installed on your computer.

Usage

  1. Firstly you should download the original model from tensorflow.
  2. Use gen_model.py to generate the train.prototxt and deploy.prototxt (or use the default prototxt).
python gen_model.py -s deploy -c 91 >deploy.prototxt
  1. Use dump_tensorflow_weights.py to dump the weights of conv layer and batchnorm layer.
  2. Use load_caffe_weights.py to load the dumped weights to deploy.caffemodel.
  3. Use the code in src to accelerate your training if you have a cudnn7, or add "engine: CAFFE" to your depthwise convolution layer to solve the memory issue.
  4. The original tensorflow model is trained on MSCOCO dataset, maybe you need deploy.caffemodel for VOC dataset, use coco2voc.py to get deploy_voc.caffemodel.

Train your own dataset

  1. Generate the trainval_lmdb and test_lmdb from your dataset.
  2. Write a labelmap.prototxt
  3. Use gen_model.py to generate some prototxt files, replace the "CLASS_NUM" with class number of your own dataset.
python gen_model.py -s train -c CLASS_NUM >train.prototxt
python gen_model.py -s test -c CLASS_NUM >test.prototxt
python gen_model.py -s deploy -c CLASS_NUM >deploy.prototxt
  1. Copy coco/solver_train.prototxt and coco/train.sh to your project and start training.

Note

There are some differences between caffe and tensorflow implementation:

  1. The padding method 'SAME' in tensorflow sometimes use the [0, 0, 1, 1] paddings, means that top=0, left=0, bottom=1, right=1 padding. In caffe, there is no parameters can be used to do that kind of padding.
  2. MobileNet on Tensorflow use ReLU6 layer y = min(max(x, 0), 6), but caffe has no ReLU6 layer. Replace ReLU6 with ReLU cause a bit accuracy drop in ssd-mobilenetv2, but very large drop in ssdlite-mobilenetv2. There is a ReLU6 layer implementation in my fork of ssd.



鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
上一篇:
kiwicom/mobile: Kiwi.com mobile app written in React Native发布时间:2022-08-29
下一篇:
ant-design/antd-mobile-samples: antd-mobile samples发布时间:2022-08-29
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap