VIRAL MIDJOURNEY/RUNWAY WORKFLOWS + THE AI VIDEO WARS

Fam,

Midjourney has been quiet for a few weeks.

This means…expect something soon (it’s inevitable)

In the meantime…we’ll talk about some more creative uses.

Let’s get into it.

Here’s what we got for today:

1. MIDJOURNEY + RUNWAY: HOME IMPROVEMENT CONCEPT

2. MIDJOURNEY + RUNWAY: SKETCH TO ANIMATION

3. AI DESIGN SYSTEMS: TEMPLATE AGENT

4. PODCAST: MJ FAST HOURS FIRST GUEST

5. BEYOND MIDJOURNEY: AI VIDEO WARS

6. MIDJOURNEY OFFICE HOURS

Let’s Boogie.

MJ + RUNWAY: HOME IMPROVEMENT CONCEPT

There’s so many examples of RW’s new Vid-2-Vid model.

But they don’t have to be complicated.

IF YOU MISSED IT → CHECK THIS SIMPLE CONCEPT.

THE CONCEPT:

How can we show a simple transformation…without significantly altering the aesthetic?

There’s a ton of creative applications for this.

(Kinda like augmented reality)

THE PROCESS:

1. Create base images in Midjourney

2. Animate base image in Runway Img-2-Vid

3. Upload vid into RW Vid-2-Vid

4. Add a text prompt

5. Generate + Iterate

MJ PROMPT EX: editorial photography of an average kitchen, full kitchen view, suburban home, bright crisp light, 35mm dslr, high precision photo --ar 16:9 --personalize [CODE] --stylize 150 --v 6.1

RW I2V PROMPT: hand-held camera shot walking through an average suburban kitchen, 35mm style

RW V2V PROMPT: hand-held camera shot walking through a Scandinavian-style kitchen, deep blue wood with gold accents, immaculate interior design, 35mm style

SIMPLE CONCEPTS SELL:

Whether you use this for a meta ad or pitching to a client….this took less than 20 minutes…and it get’s the point across.

It also works with Real iPhone Footage as well…

Check Brain Sykes work below ↓

MJ + RUNWAY: SKETCH TO ANIMATION

If you study VIRALITY on social…the “promise of a pay-off” keeps people watching.

So these sketch animations can be really useful.

IF YOU MISSED IT → FULL LINKEDIN POST

PROCESS:

1. Create sketch image with Midjourney

2. Remix (Strong) image to mockup/render

3. Remix (Strong) image to lifestyle photo.

4. Animate in Runway Gen-3.

Remix is your best friend here…because…

It will help keep the structure of the car so it translates img-2-img.

MJ Base Prompt EX: technical sketch drawing for a Tesla cyber truck, intricate detail and sketch work, cad, --ar 16:9 --personalize [code] --stylize 200 --v 6.1

MJ Remix Prompt Ex: editorial photo of a matte black Tesla cyber truck, advertising photography, dark sleek environment with pops of red, high precision photography, --ar 16:9 --personalize [code] --stylize 300 --v 6.1

ANIMATING:

What’s really impressive is Runway’s capabilities here.

This can be translated to a number of concepts.

RW I2V Prompt Ex: sketch drawing of a tesla cyber truck starting with a blank page and being created line by line

ANIMATED SKETCH WORKFLOW:

I can’t underestimate how useful this concept is…it’s easily replicable.

MJ PROMPT STRUCTURE: technical sketch drawing of the word(s) "[insert word(s)]", pencil sketch work, cad, --ar 16:9 --stylize 50 --v 6.1

RW PROMPT STRUCTURE: sketch drawing of the word "[insert word(s)]", starting with a blank page and being created line by line

I’m going to expand on this sketch idea…there’s too much potential.

Stay Tuned.

AUTOMATED AD DESIGN SYSTEM: TEMPLATE GEN

Continuing our work with Superside…this is my favorite part of the system.

Generating templates for high-performing meta ads.

It’s math and creativity blended together AND IT’S ALL DATA.

AD DESIGN SYSTEM → FULL POST HERE

AD TEMPLATES:

Look, ad templates are always in high demand…and they’re always changing. What works today, might not work tomorrow…but…visual attraction never changes. When you understand the visual focal point and what draws the eye…we can use math to structure these templates appropriately.

So when we break down ads…we turn everything into granular data.

So we can rebuild an agent that takes all these aspects into account.

TEMPLATE IMPORTANCE:

Because these agents generate off templates…our Image and Copy agents can populate multiple variations of ads..to give us options…and the ability to test.

In the meantime, if you need a CREATIVE TEAM…that’s at the cutting edge of Ai…Producing work for brands like Google, Booking.com, Coinbase, Reddit, and Meta…

PODCAST: MIDJOURNEY FAST HOURS FIRST GUEST

This week, we welcomed our first guest…Chris Branch (Midjourney OG)

Who has done AI work with likes of Coca-Cola + Frida Kahlo Estate.

»WATCH IT HERE«

HIGHLIGHTS:

• [16:17] Video AI Evolution

• [21:24] Layers in Midjourney?

• [32:37] Coca-Cola Partnership

• [42:42] Creative Ideation Process

• [47:47] Brands Embracing AI

• [52:30] AI in Hollywood

• [56:00] Monetizing AI Skills

If you like the episode…drop us a “LIKE” or “SUBSCRIBE”

Respond to this email if you have any feedback (it’s all helpful).

BEYOND MJ: THE AI VIDEO WARS

The AI video wars are heating up and there’s too much news.

So i’m doing something a little different here…

Linking to a few articles/updates below ↓

MIDJOURNEY OFFICE HOURS (9/18):

PERSONALIZATION:

1. Basic Personalization Page: Launching soon for easier access

2. Advanced Personalization: Planned for future releases

3. Multiple profiles: Style and General (personality-based)

MJ V7:

• Improved knowledge, aesthetics, prompt understanding

• Simplified architecture for more features

• Enhanced upscalers and reference capabilities

• Supports 3D and video alongside 2D images

• Potential for localized model versions

EDITOR DEVELOPMENT:

1. Standalone Editor: Separate editing from creation flow…Includes depth references and retexturing

2. Future Features: Session-based editing to avoid clutter, Advanced tools for better image manipulation, Potential drag-and-drop functionality

3D + VIDEO MODELS:

1. Video Model Progress: Improving data quality and architecture

2. Simple, cost-effective video generation this year

3. Long-term goal: real-time open-world simulation

4. Future Integrations: Possible VR applications for interactive experiences

5. Animated 3D scenes and image-to-3D capabilities

UPCOMING FEATURES:

1. Image captioning assistance tools

2. Style randomization for diverse outputs

3. 4x and 8x upscalers for better image clarity

4. Character creator tool in development

User Feedback and Surveys

• Plans to open surveys to non-Midjourney users

• Introducing polls and hive mind features

• Collecting feedback to prioritize development

Have fun.

If you need help…DM me on Linkedin.

Until next time…

Rory