-
Notifications
You must be signed in to change notification settings - Fork 30
/
index.html
75 lines (72 loc) · 4.5 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>README</title>
<link rel="stylesheet" href="https://stackedit.io/style.css" />
</head>
<body class="stackedit">
<div class="stackedit__html"><h1 id="yolov3">YoloV3</h1>
<h2 id="assignment">Assignment</h2>
<p>Assignment:</p>
<ol>
<li>OpenCV Yolo: <a href="https://pysource.com/2019/06/27/yolo-object-detection-using-opencv-with-python/">SOURCE (Links to an external site.)</a>
<ol>
<li>Run this above code on your laptop or Colab.</li>
<li>Take an image of yourself, holding another object which is there in COCO data set (search for COCO classes to learn).</li>
<li>Run this image through the code above.</li>
<li>Upload the link to GitHub implementation of this</li>
<li>Upload the annotated image by YOLO.</li>
</ol>
</li>
<li>Training Custom Dataset on Colab for YoloV3
<ol>
<li>Refer to this Colab File: <a href="https://colab.research.google.com/drive/1LbKkQf4hbIuiUHunLlvY-cc0d_sNcAgS">LINK (Links to an external site.)</a></li>
<li>Refer to this GitHub <a href="https://github.com/theschoolofai/YoloV3">Repo (Links to an external site.)</a></li>
<li>Collect a dataset of 500 images and annotate them. <strong>Please select a class for which you can find a YouTube video as well.</strong> Steps are explained in the <a href="http://readme.md">readme.md</a> file on GitHub.</li>
<li>Once done:
<ol>
<li><a href="https://www.y2mate.com/en19">Download (Links to an external site.)</a> a very small (~10-30sec) video from youtube which shows your class.</li>
<li>Use <a href="https://en.wikibooks.org/wiki/FFMPEG_An_Intermediate_Guide/image_sequence">ffmpeg (Links to an external site.)</a> to extract frames from the video.</li>
<li>Upload on your drive (alternatively you could be doing all of this on your drive to save upload time)</li>
<li>Inter on these images using <a href="http://detect.py">detect.py</a> file. <strong>Modify</strong> <a href="http://detect.py">detect.py</a> file if your file names do not match the ones mentioned on GitHub.<br>
<code>python detect.py --conf-thres 0.3 --output output_folder_name</code></li>
<li>Use <a href="https://en.wikibooks.org/wiki/FFMPEG_An_Intermediate_Guide/image_sequence">ffmpeg (Links to an external site.)</a> to convert the files in your output folder to video</li>
<li>Upload the video to YouTube.</li>
</ol>
</li>
<li>Share the link to your GitHub project with the steps as mentioned above</li>
<li>Share the link of your YouTube video</li>
</ol>
</li>
</ol>
<h1 id="yolo-v3-on-opencv">YOLO V3 on OpenCV</h1>
<ul>
<li>
<p>Colab Notebook: <a href="https://colab.research.google.com/github/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-OpenCV.ipynb">https://colab.research.google.com/github/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-OpenCV.ipynb</a></p>
</li>
<li>
<p>Github Link: <a href="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-OpenCV.ipynb">https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-OpenCV.ipynb</a></p>
</li>
</ul>
<h3 id="doggo">Doggo</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/opencv-run/doggo.png?raw=true" alt="enter image description here"></p>
<h3 id="me">Me</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/opencv-run/me.png?raw=true" alt="enter image description here"></p>
<h3 id="potto">Potto</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/opencv-run/potto.png?raw=true" alt="enter image description here"></p>
<h1 id="training-yolo-v3-on-a-custom-dataset">Training YOLO V3 on a Custom Dataset</h1>
<ul>
<li>
<p>Github Link: <a href="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-BugsBunny.ipynb">https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-BugsBunny.ipynb</a></p>
</li>
<li>
<p>Colab Notebook: <a href="https://colab.research.google.com/github/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-BugsBunny.ipynb">https://colab.research.google.com/github/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/13_YoloV3/YoloV3-BugsBunny.ipynb</a></p>
</li>
</ul>
<h1 id="running-the-custom-model-on-a-video">Running the custom model on a video</h1>
<p><a href="https://www.youtube.com/watch?v=R8sgaO6AJyE"><img src="https://img.youtube.com/vi/R8sgaO6AJyE/0.jpg" alt="BUGS BUNNY"></a></p>
</div>
</body>
</html>