Here is the entire workflow from start to finish!

Google Streetview + @runwayml References = Cinematic AI "on location". A new process for doing "on location" AI shoots. Details on how I did it below.👇

First use the image generator of your choice to create the main character. I used @midjourney 7.

Next find a location in Google Maps and then go into Streetview. Take a screenshot and then use @runwayml References with the Streetview image and the character image to do the "on location" shot. Then use Gen-4 to turn that image into a video.

For the interior shots, go to the store you want to shoot inside of in Google Maps and find a photo of inside the store you want to use and then use that interior photo with your character image.

That's pretty wild. I mean pretty soon you'll be able to make a movie in any authentic location

This is the biggest productivity cheat code right now. Kiss reading documents goodbye. You can get an instant summary of any document with this tool.

need to ‘visit’ nyc myself 😅

Insane…you could make full AI walking tours of cities 🤯 Or this would be cool for travel videos, like places to visit on a trip somewhere new.

This is insane! Imagine @Google @OfficialLoganK @demishassabis releasing a game like GTA VI, that allows us to travel anywhere in the world and play. Only Google can do it! Image to game tech, Games that can run on Cloud (Stadia), TPU's, Veo3, what not!

Thaks a ton!

Stunning work! It seems, however, that AI still has a problem with letters. But at least it can count fingers by now 😃

Brilliant and ingenious.

Thank you!

super cool, thanks!

Dozens of spelling mistakes. Couldn't even spell the stop sign properly!