1
00:00:02,480 --> 00:00:10,880
In this section, we will learn how we can integrate YOLO v9 with flask and create a web app to do object

2
00:00:10,880 --> 00:00:14,030
detection on images and on videos.

3
00:00:14,600 --> 00:00:18,290
So here is a quick demo of how our application works.

4
00:00:18,800 --> 00:00:21,230
So here you can see the upload icon.

5
00:00:21,470 --> 00:00:23,390
So I will click on this icon over here.

6
00:00:23,390 --> 00:00:25,520
And first I will upload an image.

7
00:00:25,550 --> 00:00:29,480
So let me upload this image of players play football.

8
00:00:29,540 --> 00:00:33,080
And after this I click on this play button.

9
00:00:33,080 --> 00:00:37,190
So now I will show you how we can do object detection on this image.

10
00:00:37,280 --> 00:00:44,030
So as you can click on the play button the processing starts and in two seconds you will get the output

11
00:00:44,030 --> 00:00:48,770
image where we will have the person, both of the persons as are detected.

12
00:00:48,770 --> 00:00:53,960
And you can see that, uh, this person is a bit far away and it's blurred, but our YOLO v nine model

13
00:00:53,960 --> 00:00:55,910
is able to detect this person as well.

14
00:00:55,910 --> 00:00:58,430
And you can see this person is too far away.

15
00:00:58,430 --> 00:00:59,450
Plus it is blurred.

16
00:00:59,450 --> 00:01:02,600
But our YOLO nine model has detected this person as well.

17
00:01:02,600 --> 00:01:05,780
Thus we are also able to detect sports ball as well.

18
00:01:05,780 --> 00:01:10,370
So four persons and one sports ball is being detected in this image.

19
00:01:10,640 --> 00:01:14,300
So in the same way you can upload any video as well.

20
00:01:14,300 --> 00:01:21,140
So let me upload this video in which a different persons are walking on a street.

21
00:01:21,170 --> 00:01:21,740
Okay.

22
00:01:21,740 --> 00:01:25,070
So I will click on this play button after this.

23
00:01:25,070 --> 00:01:27,110
And now you can see the processing has started.

24
00:01:27,110 --> 00:01:30,680
And uh this may take a few seconds.

25
00:01:30,680 --> 00:01:33,620
I'm running it on CPU so the processing will be very slow.

26
00:01:33,620 --> 00:01:39,530
But if you have a good chip, you just processing, uh, will be high and you will getting a good FPS.

27
00:01:39,740 --> 00:01:44,390
So now you can see that, um, all detections in this video are being done.

28
00:01:44,390 --> 00:01:45,680
Done frame by frame.

29
00:01:45,680 --> 00:01:50,210
So, uh, all the detections are being done frame by frame.

30
00:01:50,210 --> 00:01:53,180
And you can see the results over here on the screen.

31
00:01:53,180 --> 00:01:56,840
Like you can see we are able to detect the person's handbag okay.

32
00:01:56,930 --> 00:01:58,550
And the detection results are very good.

33
00:01:58,550 --> 00:02:01,100
Like you can see that this handbag is too far away.

34
00:02:01,100 --> 00:02:04,460
But our YOLO v nine model is able to detect this handbag.

35
00:02:04,460 --> 00:02:05,930
This handbag is also detected.

36
00:02:05,930 --> 00:02:10,700
But you can see that, uh, there are some persons which are very far away, but our model is successfully

37
00:02:10,700 --> 00:02:12,230
able to detect this persons.

38
00:02:12,770 --> 00:02:16,250
So we will be getting this complete web app like you can see over here.

39
00:02:16,250 --> 00:02:21,200
And we will be doing object detection or images and videos using YOLO v9.

40
00:02:21,200 --> 00:02:23,870
So let us create this web app using class.