VEO2 → Ridiculous. Nailed this abstract idea. (Inspired by Nascar's SuperBowl promo). PROMPT:.A McLaren supercar sits in a dimly lit studio, its sleek, polished body reflecting flickering footage of vintage racing history. The glossy curves capture grainy black-and-white clips

RT @Ror_Fly: Veo3 → Sloth ugc endorsements. TOOLS:.Mystic: Image Gen.Veo3: Video Gen. VEO3 PROMPT STRUCTURE:.Handheld shot slowly pushing….

Veo3 → Sloth ugc endorsements. TOOLS:.Mystic: Image Gen.Veo3: Video Gen. VEO3 PROMPT STRUCTURE:.Handheld shot slowly pushing in captures a sloth doing ugc for [PRODUCT]. The sloth rapidly holds [PRODUCT] with a snap movement. He says: ["DIALOG"]. His voice: Comical, speaking

RT @Ror_Fly: Just being the dumbest idiot. With Runway Act 2 + @dcbruck . Ref Video: Step Brothers. MoCap: Runway Act 2.Ref Image: ChatGPT….

Just being the dumbest idiot. With Runway Act 2 + @dcbruck . Ref Video: Step Brothers. MoCap: Runway Act 2.Ref Image: ChatGPT Img . (Basically, this is how MJFH started). #runwayml

RT @Ror_Fly: Runway Act II almost made me a good actor. Almost. CONTEXT:. This was a super complex test. Jack Nicholson is a maniac (….

Runway Act II almost made me a good actor. Almost. CONTEXT:. This was a super complex test. Jack Nicholson is a maniac (so good) . The hands, the facial movements, etc. Would be hard for any tool (or actor). ACT II IMPROVEMENTS:. Hand mocap is good (subtle errors). The

Midjourney Micillium Head. IMAGE PROMPT:.flawed photography, man with tiny mushrooms growing from his face, micro mist, closeup low angle shot --ar 3:4 --profile pwq76nm zrmykh6 uiav56q ggnjmiy qpfs1tj veclfmo --stylize 350. VIDEO PROMPT:.ethereal mushrooms floating upwards out

RT @Ror_Fly: Ad Creative → GPT + Runway Ref + Flora. Building usable workflows live. Looked decent on the big screen. PROCESS:.1. Revers….

Thanks for coming down the rabbit hole. More to come.

11. Flora. If you're not familiar with Flora. It's a node-based workflow builder. Where you can use:. - Multiple tools in one space.- They can be LLMs/Image Gens/Video gens.- But you can build workflows.- And share them with your team. Check it out here →

10. Continue Iteration. If you want to. you can. - Attach the final image to more image nodes.- Build out different background concepts.- Or switch it a poster style or a photography style.- Take those and turn them into video. - This becomes unlimited. Img Gen: Runway

9. Swap Ref Image + Product. Now it's a "tool" that can. - Drop in any reference image.- Analyze the composition.- Drop in your product.- And produce an image of your product .- In that composition. So it's repeatable.

8. Swap Shoes. It's a workflow that. - Can be used for other sneakers. - Just drop in a different pair. - And get the same result. - But it's not just limited to sneakers.

7. Final Product. This is a now a "workflow". - It did exactly what we wanted. - Now I can use it with one click.- But it's not just for these sneakers

6. Connecting Everything:. So we'll add. - GPT Img to our workspace. - Connect the prod image to img gen.- So it knows what to reference. Now the LLM will. - analyze the composition black shoes.- build a prompt for the composition.- And send it to the image generator. The

5. System Prompt. - Import (2) images into the workspace. - Add an LLM into our workspace as well. - Connect the ref img of the hanging black sneakers.- Build a system prompt that. - Analyzes the image.- Extracts the elements we need.- And builds the prompt. System Prompt:

4. Node Selection. For this we'll need. - Image inputs - for product/reference images.- LLM - to generate the prompts.- Image Module - To gen the images. LLM: ChatGPT 40.Img: ChatGPT Img

3. Composition Analysis. If we breakdown composition we want. - Dangling shoes.- How they're positioned.- The perspective.- The defining features. Now that we know that. We can go find out tools or nodes.

2. Image ID. - We want the composition of the shoes.- But we don't want the "black shoes".- So we ID the variables we want to control. - So how do we get composition? . (Keep reading)

1. Start with the end. First, we need to understand what we want. WE WANT:.- To put the converse shoes.- In the composition of the black shoes.- So we can reconstruct the image. Like the example below. But how do we get there?