How To Make an AI Randomizer Effect in Effect Maker with @varunraikar

0:00 Hi, I'm Bun and in this video I'll show you how to build

0:03 an AI powered randomizer effect that transforms

0:06 your selfie into a spirit animal portrait.

0:08 We'll be using the AI randomizer template in Effect Maker to do this.

0:12 By the end of this video,

0:13 you will have your own randomizer AI effect ready to publish on YouTube.

0:17 Here's how it works.

0:18 The effect cycles through different options, randomly picks one,

0:22 and based on that choice, an AI portrait is generated.

0:26 You can use this template to create personality quizzes,

0:29 random character generators, or even surprise transformations.

0:33 Basically, any effect where people will get

0:35 a different result every time they try it.

0:38 Let's dive in and customize.

0:40 First, open Effect Maker on your browser.

0:42 And here, select the AI randomizer template from the gallery.

0:45 Once you open the template, you'll see this card popup that tells you

0:49 about the effect and ways to customize it.

0:51 I'll click on Got it.

0:52 Here's your workspace.

0:53 On your left, you'll have your objects panel

0:55 where you'll see all the objects in your effect.

0:58 And here, some objects are visible and some are hidden.

1:01 You see the visible objects in the scene view.

1:04 Just beside that, we have the assets panel,

1:06 which has all the images and props neatly organized in folders.

1:11 This is where we'll customize the prompts, adjust the web generation properties,

1:14 and also update the icons to match our theme.

1:18 Below the scene view, you've got the visual script panel.

1:20 This is the brain of your effect.

1:22 Click on this X icon to see all the variables.

1:25 You can toggle the visibility of this panel

1:27 by using the change layout button on top right.

1:30 And finally, here we have the preview panel

1:32 where you can test everything in real time.

1:35 Let me show you the default effect first.

1:37 Here at the top, we see the vegetable

1:39 icon cycling and the title says vegetable portrait.

1:43 I hit record and tap the screen.

1:45 Immediately we see a flash and it stops on a vegetable.

1:49 Here the user's image is captured.

1:51 Then we get a loading screen while the AI is generating the image.

1:55 Once generated, the transformed image fades in.

1:58 And the coolest part about this is the AI

2:01 result changes based on the vegetable that got randomly selected.

2:04 This is what makes people try this effect again and again.

2:07 Now, let me show you what's happening behind the scenes.

2:10 I'll scale down the preview panel and jump to the objects panel.

2:14 Here the first object we have is the camera snapshot which

2:17 captures your selfie and sends it to the create AI image node.

2:21 This is hidden until the photo is taken

2:23 and the visual script controls when it appears.

2:26 Below that we have the background image which is currently invisible.

2:29 That is because its opacity is set to 0%.

2:32 It is just a black image whose opacity slightly increases during the loading

2:36 phase to visually tell the user that the AI is generating the image.

2:40 Next, we have the color filter, which is also hidden by default.

2:44 I will enable it and adjust the color grading here.

2:47 Watch what happens when I drag the saturation all the way to the left.

2:51 We get a black and white image.

2:53 I'll adjust this to give me a vibrant color effect.

2:56 Below that, we have a loading spinner.

2:58 It rotates while the AI is generating the image.

3:01 Its opacity currently is set to 0%.

3:03 If I increase it, you can see the loading spinner.

3:06 Let's change it to match our theme.

3:08 To change it, select the object, click on add asset,

3:11 and here upload your new loading spinner.

3:13 I made this one using the Nano Banana Pro model on the Google Gemini app.

3:17 Once uploaded, it will be automatically applied to the loading spinner object.

3:22 Next, we have icons tabbed and icon cycling.

3:24 If I enable icons tab, it's blank because its asset is

3:28 being dynamically assigned via the visual script.

3:31 This is the same for icon cycling.

3:33 It looks blank here in the scene view,

3:35 but we see the icons shuffling in the preview panel.

3:38 That is because its assets are also

3:40 being dynamically assigned in the visual script.

3:44 Next is the AI image.

3:45 This is where you'll see the transformed image appear.

3:48 It's hidden by default and only fades in once the generation completes.

3:53 Below that, we have body segmentation,

3:55 which is used to capture the user's reaction as the AI generates the image.

4:00 And finally, we have got our title text and a flash image.

4:04 The flash image is just a white image whose

4:06 opacity goes from zero to one and back to zero, mimicking a camera flash.

4:10 I'll modify the title to match our theme and I'll write,

4:13 "What's your spirit animal?" Once written,

4:16 I'll reposition it using the move tool.

4:18 Time to swap these vegetables for our cute animals.

4:21 To do that, I'll go to the assets panel,

4:23 click on upload asset, and upload my new icons.

4:27 A quick tip here is when you are creating these new icons,

4:30 check the dimensions of the original asset and make your new PGs

4:34 in the same resolution so that the new

4:36 icons fit perfectly without any adjustment.

4:39 Once uploaded, I'll open the variables panel, find the icon list variable,

4:44 and here replace each vegetable icon with my animal icons.

4:54 Now we have customized the icons, title text,

4:56 and even the color filter to match our theme.

4:59 It's time to edit the AI image prompts.

5:01 To do that, I'll go to the assets panel, and here open the prompts folder.

5:05 Set the first one, which kindly says broccoli.

5:08 I'll rename it to croc, short for crocodile.

5:10 Then I'll customize its properties.

5:15 For a full deep dive on AI image properties like character description,

5:18 post similarity, art style,

5:20 retain likeness, go watch my air transformation tutorial

5:23 where I've covered all of this in detail.

5:26 It will be linked in the description down below.

5:28 Now back to the video.

5:30 I have updated all the seven prompts to generate images that match my theme.

5:34 Now that we are done customizing this effect,

5:36 let's quickly take a look at the visual script.

5:38 The visual script has eight sections.

5:41 This is the part where the recording starts and it sets

5:44 the is recording output to true and fades out the title.

5:48 Here we have the cycling loop for icons.

5:50 This continuously shuffles the icons till the user taps.

5:54 The continuous shuffling is being handled via the frame update node.

5:59 And then here we have the main logic which checks if the user has tapped.

6:04 Once the user has tapped, it will set the tap variable to true.

6:08 and it will also disable the icon cycling and enable the icon tapped.

6:13 This icon tapped asset is dynamically set using a random

6:16 number and each number maps to a unique icon.

6:21 This random number is calculated on start using the start node

6:25 and it takes the length of the cycling icons and randomly selects

6:29 one number in that range and that same number is used

6:33 to decide which prompt should be sent to the create AI image node.

6:37 When the icons tapped is enabled, it also takes a user snapshot using the camera

6:42 snapshot node and sends it to the create image node.

6:45 Once the creation has started, this set of nodes enable the loading phase.

6:50 First, they will trigger a flash,

6:52 then enable the body segmentation and color filter.

6:54 The loading spinner is faded in along with the background.

6:58 Now once the image is generated these set

7:01 of nodes here at the bottom take the output

7:04 from create a image node and it will set

7:06 the generated image to the AI image canvas asset.

7:10 Once set the AI image canvas is also enabled.

7:13 Then its opacity increases from 0 to one to give it that fade in animation look.

7:19 As soon as it is faded in the background,

7:22 loading spinner icons tabbed and the color filter all of them are hidden.

7:26 And finally, here we have some more of these scripts.

7:29 This little section allows us to drag

7:31 the body segmentation around on the screen.

7:34 And the logic just below that rotates

7:36 the loading spinner using the frame update node.

7:39 That is the logic pro.

7:40 You don't need to customize any of this.

7:42 It just works.

7:43 And now before you submit, test it on different models in the preview panel.

7:47 Here, click on view video button and select different

7:50 models that you want to test the effect on.

8:08 Also test the effect on your phone by clicking

8:10 on the preview on phone button and scan this QR code.

8:14 Once everything works as expected,

8:16 click on submit and it will prompt you to upload a thumbnail for your effect.

8:20 For AI effects, use clear net image that shows

8:22 before on one side and after on the other.

8:25 This makes it super obvious for the user to understand what your effect does.

8:29 Click on continue and I'll name it spirit animal.

8:32 Next, for description, I'll write what the effect does.

8:35 This effect transforms your selfie into an artistic

8:39 portrait infused with your spirit animal.

8:41 And for the ticker, I'll leave it at record and tap.

8:44 Once done, I'll submit.

8:45 And that's it.

8:47 Your effect will be published to YouTube

8:48 once reviewed and approved by the YouTube team.

8:51 For more information on using Effect Maker,

8:54 check out the links in the description below.

8:55 And thank you so much for watching till the end.

8:58 Don't forget to hit the subscribe button

8:59 for more news and tutorials just like this.