Introduction

0:00 And with this new method uh that we've now got available to us, it actually unlocks a like a crazy high potential and a really useful skill that you should definitely have in your tool belt as an AI creative to potentially solve a lot of issues right now. What are these issues that we're trying to solve? Well, it's like product consistency. How do we actually like implement a product into a scene? So, let's say our starting frame that we are uploading to a Google uh like a video generator like Google V3 doesn't have the product in it.

0:22 Well, how does the camera pan to actually have the product in the scene? Well, we're going to learn how to do that today using Google V3. So, here is an example of this kind of new uh workflow template that we've got here. And you can see here, let me break this down for you. This is actually the first frame, like the first image that we're giving to the AI video generator. So, with this first image, instead of creating a photo like this one here in Midjourney that I created of this Porsche uh inside of this warehouse, let's say traditionally I uploaded this photo from MidJourney into a video generator and I said to it, transform into this a transformer.

1:09 I could describe what transformer I would want it to turn into, but again, we wouldn't have any control over that. With this new method, we can combine these elements together. So, I can say, you know, uh, the Porsche drives really quickly through the warehouse, and as it's driving, it transitions into a Optimus Prime looking white transformer robot and stands inside the warehouse, and the camera swings around uh, and looks at the transformer. So we uh so we're seeing it from the front.

1:33 Okay. So we're providing a little bit of camera motion in there. It's nothing too crazy, nothing too detailed, but we're basically telling the scene as this first frame that we're uploading to Google V3. We're basically saying this is the scene. This is what we want it to transition into. We've provided it with an annotation and a bit of text as well. This is basically our text prompt, right? So we're combining it all together into one still image frame.