1 00:00:00,280 --> 00:00:02,540 ‫Welcome to this section on machine learning. 2 00:00:02,540 --> 00:00:04,200 ‫So machine learning is not something you need to know 3 00:00:04,200 --> 00:00:06,590 ‫in-depth for the exam but there are a few 4 00:00:06,590 --> 00:00:09,400 ‫certification services that you need to know. 5 00:00:09,400 --> 00:00:12,090 ‫So the first one is Amazon Rekognition, 6 00:00:12,090 --> 00:00:14,980 ‫and as the name indicates, it's used to recognize objects, 7 00:00:14,980 --> 00:00:18,450 ‫people, text and scene in images and videos 8 00:00:18,450 --> 00:00:19,730 ‫using machine learning. 9 00:00:19,730 --> 00:00:21,920 ‫So you can do facial analysis and facial search 10 00:00:21,920 --> 00:00:24,270 ‫to do user verification and count people. 11 00:00:24,270 --> 00:00:27,160 ‫And with this, we can create a database of familiar faces 12 00:00:27,160 --> 00:00:29,410 ‫or compare against celebrities. 13 00:00:29,410 --> 00:00:31,830 ‫The use cases for Rekognition would be labeling, 14 00:00:31,830 --> 00:00:34,240 ‫content moderation, text detection, 15 00:00:34,240 --> 00:00:36,210 ‫face detection and analysis, 16 00:00:36,210 --> 00:00:39,090 ‫for example the gender, the age range and the emotions, 17 00:00:39,090 --> 00:00:42,360 ‫face search and verification, celebrity recognition, 18 00:00:42,360 --> 00:00:44,090 ‫and finally, pathing. 19 00:00:44,090 --> 00:00:46,460 ‫So if you wanna learn more, you can check out their website 20 00:00:46,460 --> 00:00:47,890 ‫which I think is very helpful. 21 00:00:47,890 --> 00:00:49,340 ‫So on the Rekognition websites, 22 00:00:49,340 --> 00:00:51,900 ‫we can we can automate our image and video analysis 23 00:00:51,900 --> 00:00:52,950 ‫with machine learning. 24 00:00:52,950 --> 00:00:53,940 ‫And I really like this website 25 00:00:53,940 --> 00:00:55,590 ‫because it shows you how it works. 26 00:00:55,590 --> 00:00:57,040 ‫So for example, for this image, 27 00:00:57,040 --> 00:00:59,130 ‫we can identify the elements of this image, 28 00:00:59,130 --> 00:01:01,570 ‫for example, a person, a rock and mountain bike, 29 00:01:01,570 --> 00:01:03,120 ‫a crest and outdoors. 30 00:01:03,120 --> 00:01:05,490 ‫We can label, for example, what we see in the images, 31 00:01:05,490 --> 00:01:07,740 ‫for example, Golden Retrievers or dogs. 32 00:01:07,740 --> 00:01:10,150 ‫Then I can look at content moderation 33 00:01:10,150 --> 00:01:13,030 ‫to make sure that it's appropriate for all ages. 34 00:01:13,030 --> 00:01:15,340 ‫I can detect text, for example for a run, 35 00:01:15,340 --> 00:01:18,480 ‫we want to see the numbers of each runner in the run. 36 00:01:18,480 --> 00:01:20,130 ‫We can do face detection and analysis. 37 00:01:20,130 --> 00:01:22,840 ‫For example, this person looks happy, she is smiling, 38 00:01:22,840 --> 00:01:24,850 ‫her eyes are open and she's a female. 39 00:01:24,850 --> 00:01:26,240 ‫Face search and verification, 40 00:01:26,240 --> 00:01:28,770 ‫maybe if you have a security application. 41 00:01:28,770 --> 00:01:30,880 ‫Finally, if you want to recognize a celebrity, 42 00:01:30,880 --> 00:01:32,280 ‫you can take a picture of them, 43 00:01:32,280 --> 00:01:34,970 ‫and this is the CTO of AWS. 44 00:01:34,970 --> 00:01:36,400 ‫And then finally, pathing. 45 00:01:36,400 --> 00:01:39,060 ‫For example, if you're monitoring a soccer game, 46 00:01:39,060 --> 00:01:41,080 ‫you could see where everyone is going to do 47 00:01:41,080 --> 00:01:43,040 ‫maybe some real time analytics. 48 00:01:43,040 --> 00:01:45,080 ‫So Rekognition is just a service you need to know 49 00:01:45,080 --> 00:01:46,550 ‫at a high level for the exam 50 00:01:46,550 --> 00:01:47,383 ‫but I really like this page 51 00:01:47,383 --> 00:01:49,430 ‫because it demonstrate the use cases. 52 00:01:49,430 --> 00:01:51,180 ‫I will see you in the next lecture.