Hong

addCodec

1 -# YOLOv4 + Deep_SORT
2 -
3 -<img src="https://github.com/yehengchen/Object-Detection-and-Tracking/blob/master/OneStage/yolo/deep_sort_yolov4/output/comparison.png" width="81%" height="81%"> <img src="https://github.com/yehengchen/video_demo/blob/master/video_demo/output.gif" width="40%" height="40%"> <img src="https://github.com/yehengchen/video_demo/blob/master/video_demo/TownCentreXVID_output.gif" width="40%" height="40%">
4 -
5 -__Object Tracking & Counting Demo - [[BiliBili]](https://www.bilibili.com/video/BV1Ug4y1i71w#reply3014975828) [[Chinese Version]](https://blog.csdn.net/weixin_38107271/article/details/96741706)__
6 -## Requirement
7 -__Development Environment: [Deep-Learning-Environment-Setup](https://github.com/yehengchen/Ubuntu-16.04-Deep-Learning-Environment-Setup)__
8 -
9 -* OpenCV
10 -* sklean
11 -* pillow
12 -* numpy 1.15.0
13 -* torch 1.3.0
14 -* tensorflow-gpu 1.13.1
15 -* CUDA 10.0
16 -***
17 -
18 -It uses:
19 -
20 -* __Detection__: [YOLOv4](https://github.com/yehengchen/Object-Detection-and-Tracking/tree/master/OneStage/yolo/Train-a-YOLOv4-model) to detect objects on each of the video frames. - 用自己的数据训练YOLOv4模型
21 -
22 -* __Tracking__: [Deep_SORT](https://github.com/nwojke/deep_sort) to track those objects over different frames.
23 -
24 -*This repository contains code for Simple Online and Realtime Tracking with a Deep Association Metric (Deep SORT). We extend the original SORT algorithm to integrate appearance information based on a deep appearance descriptor. See the [arXiv preprint](https://arxiv.org/abs/1703.07402) for more information.*
25 -
26 -## Quick Start
27 -
28 -__0.Requirements__
29 -
30 - pip install -r requirements.txt
31 -
32 -__1. Download the code to your computer.__
33 -
34 - git clone https://github.com/yehengchen/Object-Detection-and-Tracking.git
35 -
36 -__2. Download [[yolov4.weights]](https://drive.google.com/file/d/1cewMfusmPjYWbrnuJRuKhPMwRe_b9PaT/view) [[Baidu]](https://pan.baidu.com/s/1jRudrrXAS3DRGqT6mL4L3A ) - `mnv6`__ and place it in `deep_sort_yolov4/model_data/`
37 -
38 -*Here you can download my trained [[yolo4_weight.h5]](https://pan.baidu.com/s/1JuT4KCUFaE2Gvme0_S37DQ ) - `w17w` weights for detecting person/car/bicycle,etc.*
39 -
40 -__3. Convert the Darknet YOLO model to a Keras model:__
41 -```
42 -$ python convert.py model_data/yolov4.cfg model_data/yolov4.weights model_data/yolo.h5
43 -```
44 -__4. Run the YOLO_DEEP_SORT:__
45 -
46 -```
47 -$ python main.py -c [CLASS NAME] -i [INPUT VIDEO PATH]
48 -
49 -$ python main.py -c person -i ./test_video/testvideo.avi
50 -```
51 -
52 -__5. Can change [deep_sort_yolov3/yolo.py] `__Line 100__` to your tracking object__
53 -
54 -*DeepSORT pre-trained weights using people-ReID datasets only for person*
55 -```
56 - if predicted_class != args["class"]:
57 - continue
58 -
59 - if predicted_class != 'person' and predicted_class != 'car':
60 - continue
61 -```
62 -
63 -## Train on Market1501 & MARS
64 -*People Re-identification model*
65 -
66 -[cosine_metric_learning](https://github.com/nwojke/cosine_metric_learning) for training a metric feature representation to be used with the deep_sort tracker.
67 -
68 -## Citation
69 -
70 -### YOLOv4 :
71 -
72 - @misc{bochkovskiy2020yolov4,
73 - title={YOLOv4: Optimal Speed and Accuracy of Object Detection},
74 - author={Alexey Bochkovskiy and Chien-Yao Wang and Hong-Yuan Mark Liao},
75 - year={2020},
76 - eprint={2004.10934},
77 - archivePrefix={arXiv},
78 - primaryClass={cs.CV}
79 - }
80 -
81 -### Deep_SORT :
82 -
83 - @inproceedings{Wojke2017simple,
84 - title={Simple Online and Realtime Tracking with a Deep Association Metric},
85 - author={Wojke, Nicolai and Bewley, Alex and Paulus, Dietrich},
86 - booktitle={2017 IEEE International Conference on Image Processing (ICIP)},
87 - year={2017},
88 - pages={3645--3649},
89 - organization={IEEE},
90 - doi={10.1109/ICIP.2017.8296962}
91 - }
92 -
93 - @inproceedings{Wojke2018deep,
94 - title={Deep Cosine Metric Learning for Person Re-identification},
95 - author={Wojke, Nicolai and Bewley, Alex},
96 - booktitle={2018 IEEE Winter Conference on Applications of Computer Vision (WACV)},
97 - year={2018},
98 - pages={748--756},
99 - organization={IEEE},
100 - doi={10.1109/WACV.2018.00087}
101 - }
102 -
103 -## Reference
104 -#### Github:deep_sort@[Nicolai Wojke nwojke](https://github.com/nwojke/deep_sort)
105 -#### Github:deep_sort_yolov3@[Qidian213 ](https://github.com/Qidian213/deep_sort_yolov3)
106 -#### Github:Deep-SORT-YOLOv4@[LeonLok](https://github.com/LeonLok/Deep-SORT-YOLOv4)
107 -
No preview for this file type