代码运行方法 deep sort yolov3： real-time Multi-person tracker using YOLO v3 and deep_sort with tensorflow

github网址：

一：

https://github.com/nwojke/deep_sort   #这个是论文代码deep sort

此论文的检测部分采取的是下面的这篇论文参数  ECCV2016  （https://blog.csdn.net/sunshinezhihuo/article/details/78885012）

SORT：

论文地址：http://arxiv.org/pdf/1602.00763.pdf

代码地址：https://github.com/abewley/sort

deep-SORT：

论文地址：https://arxiv.org/pdf/1703.07402.pdf

代码链接：https://github.com/nwojke/deep_sort
 

二：

https://github.com/qqwweee/keras-yolo3   #这是YOLO的检测部分

训练的博客：https://blog.csdn.net/Patrick_Lxc/article/details/80615433

三：

https://github.com/Qidian213/deep_sort_yolov3    #此作者用的YOLO

视频网址：

https://www.bilibili.com/video/av23500163/

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Introduction

Thanks for these projects, this work now is support tiny_yolo v3 but only for test, if you want to train you can either train a model in darknet or in the second following works. It also can tracks many objects in coco classes, so please note to modify the classes in yolo.py. besides, you also can use camera for testing.

https://github.com/nwojke/deep_sort           #这里是deep_sort的代码

https://github.com/qqwweee/keras-yolo3      #这里是keras_yolo3的代码

https://github.com/Qidian213/deep_sort_yolov3     #这里是deep_sort结合yolov3的代码

 

Quick Start

1. Download YOLOv3 or tiny_yolov3 weights from YOLO website.Then convert the Darknet YOLO model to a Keras model. Or use what i had converted https://drive.google.com/file/d/1uvXFacPnrSMw6ldWTyLLjGLETlEsUvcE/view?usp=sharing (yolo.h5 model file with tf-1.4.0) , put it into model_data folder
   ＃下载权重从Darknet YOLO model 转换为 a Keras model　yolo.h5，也可用现成的放入model_data
   

2. Run YOLO_DEEP_SORT with cmd :
   ＃终端运行

   python demo.py
   

3. (Optional) Convert the Darknet YOLO model to a Keras model by yourself:
   ＃自己转化，权重从Darknet YOLO model 转换为 a Keras model

 please download the weights at first from yolo website. 
 python convert.py yolov3.cfg yolov3.weights model_data/yolo.h5

 

Dependencies

The code is compatible with Python 2.7 and 3. The following dependencies are needed to run the tracker:

NumPy
sklean
OpenCV

Additionally, feature generation requires TensorFlow-1.4.0.

 

Note

Model file model_data/mars-small128.pb need by deep_sort had convert to tensorflow-1.4.0

＃作者已经放进去了

 

Test only

Speed : when only run yolo detection about 11-13 fps , after add deep_sort about 11.5 fps (GTX1060 6G)

Test result video : https://www.bilibili.com/video/av23500163/ generated by this project

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

问题一：

自己转换的过程，利用convert.py文件转换为yolo.h5文件

（我这样做的原因是：作者github上需要tf1.4.0,我想这个自己转换是不是就可以适用于我的1.6.0版本，可是这样真的不行)

python convert.py yolov3.cfg yolov3.weights model_data/yolo.h5

此步会报错，'convet.py'TypeError:a bytes-like object is required ,not str 

报错如下

像20行这样改就可以了

问题二：

在spyder里面运行demo.py不行，在终端下，python demo.py这样可以运行

应该是我的ＩＤＥ问题