1 00:00:02,000 --> 00:00:07,000 Now you can see over here, I have opened the pigeon, so I will just click over here, file, click 2 00:00:07,000 --> 00:00:09,000 on new project from here. 3 00:00:09,000 --> 00:00:10,000 Okay. 4 00:00:10,000 --> 00:00:18,000 So I will just now redirect towards the directory where I am just created a folder so it's inside the. 5 00:00:19,000 --> 00:00:25,000 Okay, so now I'm inside the directory and here I can see the folder by the name YOLO V8 Crash course. 6 00:00:25,000 --> 00:00:28,000 So this is the folder I just created for this tutorial. 7 00:00:29,000 --> 00:00:30,000 So just click over here. 8 00:00:31,000 --> 00:00:31,000 Okay. 9 00:00:31,000 --> 00:00:34,000 Then I'm just creating a new environment. 10 00:00:34,000 --> 00:00:40,000 You can see over here I'm just creating a new virtual environment over here and this is my base interpreter. 11 00:00:40,000 --> 00:00:41,000 Python dot exe. 12 00:00:41,000 --> 00:00:44,000 Okay, so just click over here. 13 00:00:45,000 --> 00:00:50,000 So this is my base interpreter and then click on Create New project. 14 00:00:54,000 --> 00:00:57,000 As I click on Create, it will ask me. 15 00:00:57,000 --> 00:00:59,000 Projects can either be open a new window or replace the project. 16 00:00:59,000 --> 00:01:02,000 So just click on this window from here. 17 00:01:02,000 --> 00:01:05,000 Okay, so now it's creating a new virtual environment. 18 00:01:05,000 --> 00:01:11,000 You can see over here and this is our directory name, YOLO V8 Crash course, which you can see over 19 00:01:11,000 --> 00:01:11,000 here. 20 00:01:11,000 --> 00:01:17,000 Now we can see that the virtual environment is being created, so this might take few minutes as it's 21 00:01:17,000 --> 00:01:20,000 updating the interpreter paths and everything. 22 00:01:20,000 --> 00:01:26,000 Okay, so now you can see that we have created a new project over here and here we have the main.py 23 00:01:27,000 --> 00:01:27,000 file. 24 00:01:27,000 --> 00:01:32,000 So if you just click over here, just right click over here and then just go to. 25 00:01:34,000 --> 00:01:34,000 From here. 26 00:01:40,000 --> 00:01:40,000 Sorry. 27 00:01:40,000 --> 00:01:41,000 Just click over here. 28 00:01:41,000 --> 00:01:41,000 Fine. 29 00:01:41,000 --> 00:01:43,000 And then click on from here. 30 00:01:43,000 --> 00:01:48,000 Just click on settings from here and then just click on over here. 31 00:01:48,000 --> 00:01:54,000 Project Yolo V eight Crash Course is the name of our project, and then then select click on Python 32 00:01:54,000 --> 00:01:54,000 Interpreter. 33 00:01:54,000 --> 00:01:55,000 Interpreter. 34 00:01:55,000 --> 00:01:59,000 So now you can see that these three packages are already installed. 35 00:01:59,000 --> 00:02:06,000 So now you can see that we have downloaded Python 3.10, so we can see Python 3.10 over here and YOLO 36 00:02:06,000 --> 00:02:07,000 V8 crash course. 37 00:02:07,000 --> 00:02:13,000 And so just install to install any new package, just click over plus and just write here. 38 00:02:13,000 --> 00:02:19,000 For example, if I just want to install numpy, I will just type numpy and just select numpy and click 39 00:02:19,000 --> 00:02:20,000 on install package. 40 00:02:20,000 --> 00:02:21,000 Okay. 41 00:02:21,000 --> 00:02:23,000 So it will install the package. 42 00:02:23,000 --> 00:02:29,000 So in this way you can just write all the packages name over here and just install all those packages 43 00:02:29,000 --> 00:02:30,000 like you can see over here. 44 00:02:31,000 --> 00:02:37,000 If I just want to install Cv2 package Cv2 package I will just write OpenCV python. 45 00:02:39,000 --> 00:02:46,000 So that's right OpenCV Python So you can see that here we have the dot package. 46 00:02:46,000 --> 00:02:51,000 So if you just click over here and you can install the Cv2 OpenCV Python package as well. 47 00:02:52,000 --> 00:02:54,000 But this process takes very much time. 48 00:02:54,000 --> 00:03:00,000 Like in this way you just need to write every single package name over here in the search bar and install 49 00:03:00,000 --> 00:03:01,000 that package. 50 00:03:01,000 --> 00:03:04,000 There is another, another way will be also to install the package. 51 00:03:04,000 --> 00:03:06,000 So let me just tell you. 52 00:03:06,000 --> 00:03:12,000 So now you can see that as I install the numpy package, we can see that numpy package appears over 53 00:03:12,000 --> 00:03:15,000 here and now we have the four packages installed. 54 00:03:15,000 --> 00:03:20,000 In this way you can just click over here plus and install any of the package which you require. 55 00:03:20,000 --> 00:03:23,000 If so, just going back from here. 56 00:03:24,000 --> 00:03:30,000 So in order to install the packages of your choice, you just need to click over here, file and create 57 00:03:30,000 --> 00:03:33,000 a file by the name requirements dot txt. 58 00:03:33,000 --> 00:03:38,000 Okay, so in this file you can write all the packages which you want to install. 59 00:03:38,000 --> 00:03:46,000 So if I just want to install ultralytics, I will just write ultralight 8.0.26. 60 00:03:47,000 --> 00:03:47,000 Okay. 61 00:03:47,000 --> 00:03:48,000 So this is a version. 62 00:03:48,000 --> 00:03:49,000 Okay. 63 00:03:49,000 --> 00:03:53,000 In this way if you want to install numpy package, although I've just already installed, you can just 64 00:03:53,000 --> 00:03:59,000 write numpy and if you just want to install matplotlib you can just write the name of package over here. 65 00:03:59,000 --> 00:04:02,000 So now here you can see over here. 66 00:04:02,000 --> 00:04:05,000 Package requirement Ultralytics Matplotlib are not satisfied. 67 00:04:05,000 --> 00:04:08,000 So if you just click on install requirements. 68 00:04:08,000 --> 00:04:13,000 So now it will just install all these two packages and just click on install from here. 69 00:04:13,000 --> 00:04:19,000 So now you can see that just go below down over here, you can see that installing package. 70 00:04:19,000 --> 00:04:26,000 So it's installing the Ultralytics package and after this it will install the Matplotlib package, although 71 00:04:26,000 --> 00:04:28,000 numpy package is already installed. 72 00:04:28,000 --> 00:04:28,000 Okay. 73 00:04:28,000 --> 00:04:29,000 So. 74 00:04:30,000 --> 00:04:35,000 But just let's run the Main.py file and see what output do we get from here. 75 00:04:35,000 --> 00:04:41,000 So if I just click on Main.py file and just click on Run Main and just open the terminal so you can 76 00:04:41,000 --> 00:04:43,000 see that it's giving me the output. 77 00:04:43,000 --> 00:04:44,000 Hi pi job. 78 00:04:44,000 --> 00:04:46,000 So that's fine. 79 00:04:50,000 --> 00:04:53,000 In the meanwhile that alter prefix package get installed. 80 00:04:53,000 --> 00:04:59,000 And let me tell you in this video tutorial I will be using YOLO V8 for the object detection. 81 00:04:59,000 --> 00:05:00,000 So alternate. 82 00:05:01,000 --> 00:05:07,000 So to implement YOLO a8 we just need to install the ultralytics package which I'm just installing over 83 00:05:07,000 --> 00:05:07,000 here. 84 00:05:07,000 --> 00:05:11,000 So YOLO eight is the latest release of the YOLO version. 85 00:05:11,000 --> 00:05:18,000 So YOLO V8 outperforms all the other YOLO version in terms of accuracy and speed and the mean average 86 00:05:18,000 --> 00:05:22,000 precision of 56 7.6 have has been marked with YOLO V8. 87 00:05:23,000 --> 00:05:26,000 So if I just show you the YOLO V8 GitHub repository. 88 00:05:26,000 --> 00:05:28,000 So let me just show you. 89 00:05:28,000 --> 00:05:36,000 So if you just click over here and just write YOLO V8 it up, Okay, So and just click on this first 90 00:05:36,000 --> 00:05:36,000 link. 91 00:05:37,000 --> 00:05:44,000 So Ultralytics has released YOLO V8 Ultralytics has also released YOLO V5 and YOLO V3 so you'll be eight 92 00:05:44,000 --> 00:05:48,000 is the only version of YOLO which has its own package. 93 00:05:48,000 --> 00:05:54,000 So you just need to install PIP, install ultralytics and you can just run YOLO V8. 94 00:05:54,000 --> 00:06:00,000 So YOLO V8 is the only version of YOLO which has its own package that you can see over here. 95 00:06:00,000 --> 00:06:04,000 Pip install Ultralytics will just install the YOLO V8 on your. 96 00:06:05,000 --> 00:06:09,000 System and you just need to write YOLO predict and you can just run YOLO V8. 97 00:06:09,000 --> 00:06:15,000 Okay, so prior versions of YOLO had don't have their own package, but YOLO V8 had its own package 98 00:06:15,000 --> 00:06:18,000 which is ultra, which is PIP install already. 99 00:06:18,000 --> 00:06:21,000 So using PIP install ultralytics you can just install YOLO v8. 100 00:06:22,000 --> 00:06:27,000 You can also, if you want to make any change in the detection or training script, you can just clone 101 00:06:27,000 --> 00:06:29,000 the repo as well. 102 00:06:29,000 --> 00:06:29,000 Okay. 103 00:06:29,000 --> 00:06:37,000 So in our case we are just installing the YOLO V8 package by just writing ultralytics and its version. 104 00:06:37,000 --> 00:06:37,000 Okay. 105 00:06:37,000 --> 00:06:39,000 In the requirements.txt file. 106 00:06:39,000 --> 00:06:46,000 So if I just go back over here now you can see over here in the requirements.txt file I'm just installing 107 00:06:46,000 --> 00:06:49,000 YOLO V8 by just writing Ultralytics and its version. 108 00:06:49,000 --> 00:06:52,000 So in this way we can install the YOLO V8. 109 00:06:52,000 --> 00:06:55,000 So now we can see that package is successfully installed. 110 00:06:55,000 --> 00:06:59,000 Packages ultralytics and Matplotlib package has been successfully installed. 111 00:06:59,000 --> 00:07:06,000 You can see in the node over here we go down like the package is installed successfully, which includes 112 00:07:06,000 --> 00:07:13,000 Ultralytics and Matplotlib package and you can see that we are using Python 3.10, which you can see 113 00:07:13,000 --> 00:07:20,000 over in the top in the bottom right corner, Python 3.10, which we are using, which we have just downloaded 114 00:07:20,000 --> 00:07:25,000 the python 3.10 version from python.org. 115 00:07:25,000 --> 00:07:26,000 Okay. 116 00:07:26,000 --> 00:07:28,000 So now we will just run YOLO V8. 117 00:07:28,000 --> 00:07:36,000 So first of all I will just go over here and click on new and create a directory by the name run YOLO 118 00:07:36,000 --> 00:07:39,000 by running YOLO. 119 00:07:40,000 --> 00:07:41,000 Be it. 120 00:07:41,000 --> 00:07:44,000 Okay, so this is our directory. 121 00:07:44,000 --> 00:07:45,000 Running YOLO V8. 122 00:07:45,000 --> 00:07:47,000 And here I will just create a file by the name. 123 00:07:49,000 --> 00:07:49,000 Your law. 124 00:07:53,000 --> 00:07:54,000 death.py. 125 00:07:54,000 --> 00:07:55,000 Okay. 126 00:07:57,000 --> 00:08:00,000 So now you can see that as we have installed the AllPolitics package. 127 00:08:00,000 --> 00:08:02,000 So I will just write from Ultralytics. 128 00:08:05,000 --> 00:08:07,000 Import yolo. 129 00:08:07,000 --> 00:08:13,000 Okay, so it will import yolo v8 from you by using from ultralytics import yolo. 130 00:08:13,000 --> 00:08:16,000 It will automatically import the yolo V8. 131 00:08:17,000 --> 00:08:17,000 Okay. 132 00:08:17,000 --> 00:08:19,000 Yolo v8 package from ultralytics. 133 00:08:19,000 --> 00:08:21,000 So in the next step I'll just write. 134 00:08:24,000 --> 00:08:25,000 Import cv2. 135 00:08:25,000 --> 00:08:28,000 Or let's just skip it and just write. 136 00:08:28,000 --> 00:08:29,000 Model is equal to. 137 00:08:30,000 --> 00:08:32,000 Then we just need to write YOLO. 138 00:08:34,000 --> 00:08:35,000 And just right. 139 00:08:36,000 --> 00:08:40,000 So in case of Angola, we have different. 140 00:08:42,000 --> 00:08:43,000 Evidence available. 141 00:08:43,000 --> 00:08:44,000 So let me just show you. 142 00:08:44,000 --> 00:08:47,000 So now if I just go below down here. 143 00:08:47,000 --> 00:08:52,000 So now you can see that we have five different weights, Yolo V eight and or Nano is the smallest and 144 00:08:52,000 --> 00:08:57,000 it is it is the smallest in size and it is the fastest as well. 145 00:08:57,000 --> 00:09:05,000 But it is less accurate while yolo V8X is the is the more accurate, but it is less fast or it is slow 146 00:09:05,000 --> 00:09:07,000 as compared to other YOLO V8 versions. 147 00:09:07,000 --> 00:09:14,000 So simply yolo v eight and is less accurate, but it is the fastest but yolo V8X is more accurate than 148 00:09:14,000 --> 00:09:16,000 other all other yolo v eight. 149 00:09:16,000 --> 00:09:18,000 But it is less fast. 150 00:09:18,000 --> 00:09:18,000 Okay. 151 00:09:19,000 --> 00:09:23,000 So just go back from here and just write your logo. 152 00:09:24,000 --> 00:09:26,000 B a n dot p t, Okay. 153 00:09:26,000 --> 00:09:32,000 So I'm just writing the name of the weight file over here so the weight will be automatically downloaded. 154 00:09:32,000 --> 00:09:32,000 Okay. 155 00:09:32,000 --> 00:09:37,000 So just need to we don't need to manually download the weight and just place in this folder. 156 00:09:37,000 --> 00:09:39,000 The weights will be automatically downloaded. 157 00:09:39,000 --> 00:09:44,000 You just need to write the name of the weights which you want to use. 158 00:09:44,000 --> 00:09:46,000 Just need to write the name of the weights. 159 00:09:46,000 --> 00:09:47,000 Pre-trained weights over here. 160 00:09:47,000 --> 00:09:48,000 Okay. 161 00:09:48,000 --> 00:09:56,000 And then after this, just write results is equal to model and just pass the input image path over here. 162 00:09:56,000 --> 00:09:56,000 Okay. 163 00:09:56,000 --> 00:09:58,000 So let me just. 164 00:10:00,000 --> 00:10:01,000 Okay. 165 00:10:01,000 --> 00:10:04,000 So let me just add some image over here. 166 00:10:04,000 --> 00:10:10,000 So have already have some downloaded some images from, uh, Google. 167 00:10:10,000 --> 00:10:12,000 So I will just add these images over here. 168 00:10:12,000 --> 00:10:18,000 So let me just go back and just let me some add some new images. 169 00:10:18,000 --> 00:10:18,000 Okay? 170 00:10:18,000 --> 00:10:21,000 So just just give me a minute. 171 00:10:21,000 --> 00:10:24,000 I'm just going back and just adding some new images. 172 00:10:25,000 --> 00:10:25,000 Okay. 173 00:10:25,000 --> 00:10:28,000 So let me just show you as well. 174 00:10:30,000 --> 00:10:33,000 Well, if I just go over here in this folder. 175 00:10:33,000 --> 00:10:38,000 So now in the images folder, you can see over here, I have just placed these four images. 176 00:10:38,000 --> 00:10:44,000 Okay, So let me just test out on these images and see what results do we get. 177 00:10:44,000 --> 00:10:44,000 Okay. 178 00:10:47,000 --> 00:10:51,000 So I'm just going back to our stock code and just I will just write down, down. 179 00:10:51,000 --> 00:10:52,000 So I will. 180 00:10:53,000 --> 00:10:57,000 Now I just want to go to the images folder and just write the name of the image or the image. 181 00:10:58,000 --> 00:10:59,000 I just need to add the image path. 182 00:10:59,000 --> 00:11:05,000 So I will just write, uh, dot dot, which means go backward, like go outside this running Yolo V 183 00:11:05,000 --> 00:11:12,000 folder and then just go to images folder and just write, uh, the three dot PNG, the name of the image 184 00:11:12,000 --> 00:11:16,000 on which I just want to perform the detection. 185 00:11:16,000 --> 00:11:16,000 Okay. 186 00:11:16,000 --> 00:11:20,000 And just write show is equal to true. 187 00:11:20,000 --> 00:11:24,000 Okay, so now just click on over here. 188 00:11:24,000 --> 00:11:26,000 yolo-test.py. 189 00:11:26,000 --> 00:11:30,000 So just click over here and just run YOLO dash test this script. 190 00:11:30,000 --> 00:11:33,000 So I'm just running this YOLO test.py script. 191 00:11:33,000 --> 00:11:36,000 Okay, so this might take few seconds. 192 00:11:46,000 --> 00:11:52,000 Now you can see that the output just appeared and it just flows because we don't have added any delay. 193 00:11:52,000 --> 00:11:52,000 Okay. 194 00:11:52,000 --> 00:11:54,000 So that's why we cannot see the output. 195 00:11:54,000 --> 00:11:56,000 It just appeared and went away. 196 00:11:56,000 --> 00:12:01,000 So because we don't have added we because we haven't added any delay. 197 00:12:01,000 --> 00:12:02,000 So let's add the delay. 198 00:12:02,000 --> 00:12:05,000 So to add the delay, I will just import cv2. 199 00:12:09,000 --> 00:12:12,000 Import cv2 and I will just write. 200 00:12:13,000 --> 00:12:15,000 CV to dot Wait key. 201 00:12:16,000 --> 00:12:19,000 So I'm just adding that delay now and just write zero over here. 202 00:12:19,000 --> 00:12:22,000 And now let's run this script again. 203 00:12:22,000 --> 00:12:25,000 So now hopefully we will have the output in front of us. 204 00:12:27,000 --> 00:12:28,000 Okay. 205 00:12:29,000 --> 00:12:30,000 The USB. 206 00:12:30,000 --> 00:12:33,000 We can't see the output because we don't have any added any delay. 207 00:12:33,000 --> 00:12:35,000 So. 208 00:12:40,000 --> 00:12:43,000 So now you can see that we have the output in front of us. 209 00:12:43,000 --> 00:12:46,000 Like you can see that this is the input image I have just passed. 210 00:12:46,000 --> 00:12:50,000 And here we have the detection results using YOLO V8. 211 00:12:50,000 --> 00:12:53,000 So now you can see that the model is able to detect motorcycles. 212 00:12:53,000 --> 00:12:59,000 Like you can see that we have multiple motorcycles plus in the back side of the image we have the cars. 213 00:12:59,000 --> 00:13:02,000 So the model is able to detect cars as well as the motorcycle. 214 00:13:02,000 --> 00:13:04,000 So the detection results are quite fine. 215 00:13:04,000 --> 00:13:09,000 And let's test on some other images and well and see what results do we get. 216 00:13:09,000 --> 00:13:12,000 Okay, so just write the name. 217 00:13:12,000 --> 00:13:16,000 So I'm just passing two dot PNG and see what results do we get. 218 00:13:16,000 --> 00:13:23,000 So just click over here and just run YOLO test and okay, just forgot to stop the previous. 219 00:13:23,000 --> 00:13:30,000 So I will just click over here and stop the previous running script and just run this again now. 220 00:13:31,000 --> 00:13:33,000 So run YOLO test. 221 00:13:33,000 --> 00:13:34,000 Okay. 222 00:13:38,000 --> 00:13:39,000 Okay, so let's see. 223 00:13:39,000 --> 00:13:41,000 What results do we get in this case from here? 224 00:13:42,000 --> 00:13:45,000 Okay, so let me just share my screen now. 225 00:13:46,000 --> 00:13:51,000 So now you can see over here, like you can see over here, we have detected the card. 226 00:13:51,000 --> 00:13:52,000 This is a wrong detection. 227 00:13:52,000 --> 00:13:53,000 This is not a bus. 228 00:13:53,000 --> 00:13:55,000 This is also a card file. 229 00:13:55,000 --> 00:14:00,000 You can see that the YOLO V8 model is detecting cards and these are very impressive results. 230 00:14:00,000 --> 00:14:08,000 Currently we are using nano, but I have already told you that YOLO is the largest among the YOLO V8 231 00:14:08,000 --> 00:14:10,000 model and it is more accurate. 232 00:14:10,000 --> 00:14:17,000 So you can use also use the YOLO V8 weights and in that if you use the YOLO V8 weights, the detection 233 00:14:17,000 --> 00:14:23,000 results will be more backward because YOLO, A10 Nano is also the fastest, but it is less accurate 234 00:14:23,000 --> 00:14:27,000 while YOLO V8 is less fast but is more accurate. 235 00:14:27,000 --> 00:14:29,000 Okay, so the detection results are fine. 236 00:14:29,000 --> 00:14:31,000 So let me just go back. 237 00:14:32,000 --> 00:14:36,000 Okay, so now what we can do is. 238 00:14:37,000 --> 00:14:40,000 You can test on some other images as well. 239 00:14:40,000 --> 00:14:46,000 But I I've already tested on two images two dot PNG and three dot PNG already.