Giter Site home page Giter Site logo

xyt2008 / caffe-yolov2-windows Goto Github PK

View Code? Open in Web Editor NEW

This project forked from eric612/caffe-yolov3-windows

0.0 2.0 0.0 255.71 MB

A windows caffe implementation of YOLOv2 detection network, with pretrained weights on VOC0712 and mAP=0.709

License: Other

CMake 3.97% Makefile 0.58% Shell 0.42% PowerShell 0.02% Dockerfile 0.07% C++ 79.65% Cuda 6.13% MATLAB 0.78% Python 8.18% Batchfile 0.20%

caffe-yolov2-windows's Introduction

caffe-yolov2-windows

A caffe implementation of MobileNet-YOLO (YOLOv2 base) detection network, with pretrained weights on VOC0712 and mAP=0.709

Network mAP Download Download NetScope
MobileNet-YOLO-Lite 0.675 train deploy graph
MobileNet-YOLO 0.709 train deploy graph

Note : Training from linux version and test on windows version , the mAP of MobileNetYOLO-lite was 0.668

Performance

Compare with YOLOv2

Network mAP Weight size Inference time (GTX 1080)
MobileNet-YOLO-Lite 0.675 16.8 mb 10 ms
MobileNet-YOLO 0.709 19.4 mb 24 ms
Tiny-YOLO 0.57 60.5 mb N/A
YOLOv2 0.76 193 mb N/A

Note : the yolo_detection_output_layer not be optimization , and batch norm and scale layer can merge into conv layer

Linux Version

MobileNet-YOLO

Modifications

  1. caffe training
  2. add pre-trained model
  3. fix bugs
  4. windows support

Configuring and Building Caffe

Requirements

  • Visual Studio 2013 or 2015
  • CMake 3.4 or higher (Visual Studio and Ninja generators are supported)
  • Anaconda

The build step was the same as MobileNet-SSD-windows

> cd $caffe_root
> script/build_win.cmd 

Darknet YOLOv2 Demo (COCO)

Download yolov2 coco weights

Save at $caffe_root/models/convert

cd $caffe_root/models/convert

python weights_to_prototxt.py

Note : you may need specify python caffe path or copy python lib. here

cd $caffe_root

examples\demo_darknet19.cmd

MobilenetYOLO Demo

Download deploy model

Save at $caffe_root/models/yolov2

> cd $caffe_root/
> examples\demo_yolo_lite.cmd

If load success , you can see the image window like this

alt tag

Trainning Prepare

Download lmdb

Unzip into $caffe_root/

Please check the path exist "$caffe_root\examples\VOC0712\VOC0712_trainval_lmdb"

Download pre-trained weights , and save at $caffe_root\model\convert

Training Darknet YOLOv2

> cd $caffe_root/
> examples\train_yolo_darknet.cmd

Trainning MobilenetYOLO

> cd $caffe_root/
> examples\train_yolo.cmd

Future work

  1. yolov3

Reference

https://github.com/eric612/Vehicle-Detection

https://github.com/eric612/MobileNet-SSD-windows

https://github.com/gklz1982/caffe-yolov2

https://github.com/duangenquan/YoloV2NCS

caffe-yolov2-windows's People

Contributors

eric612 avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.