Artificial intelligence-classify images

Artificial Intelligence
1

Hi! I've done the app explained in the video https://www.youtube.com/watch?v=kd0_6DfwQDQ

I've followed all the steps but my app classifies all the objects incorrectly, even simple objects (a pencil, the mouse...). Any ideas why this happens? Thank you

2

OK. Perhaps there is an error in your code. We can not tell because you did not share your Blocks. :cry:

Here is a link to a tutorial regarding that Project ( I think) Introduction to Machine Learning: Image Classification

Confirm your Blocks look like the Blocks in the article and that you loaded the Look extension correctly.

Perhaps you have not trained enough objects?

3

Thanks for your answer Steve. The code is correct. Do I have to train it? How? I thought that was not necessary. I see that my phone (a redmi 9 note ) is not on the list of compatible devices for AI extensions, I don't know if that is the problem. I've also tried the app on my chromebook but it does not work either (besides how you manage to focus the webcam on the object?). The third option, installing the MIT app on the chromebook does not work either (when I write the code of the app nothing happens)

4

  1. your hardware might be incompatible NOTE: Not all mobile devices/operating systems currently have the required hardware/software to run the Look extension used in this unit. Please check here to see if your mobile device is on our list of devices where the extension is known to work. If your device is not on the list, we highly recommend testing beforehand to make sure it is compatible.

  2. Chromebook a Chromebook, as most will support the LookExtension. In addition, many Chromebooks now support running Android apps, so the MIT AI2 Companion can be installed directly on the Chromebook, and students can run the App Inventor IDE and the AI2 Companion on the same machine.

  3. Train. Yes you have to train it. Train a Model with the Personal Image Classifier
    Students open Web browsers on their computers and use http://classifier.appinventor.mit.edu/oldpic/ to create a model of different facial expressions.

If you have not trained the PIC might be the reason

I don't have a Chromebook .

I can tell you the advice to use the Look extension does work on my Android 8.1 Samsung tablet. A guess is it might not work on your redmi because of hardware issues or not training . Since you haven't trained, you may have to do that to get either your redmi or chromebook to work.

Train the software and try again?

5

Can you add a project link if you can?

6

Scroll to the end of page and go to tutorial

image