When working on a project, you can find a style and color grade that you like, a film texture, and lock it into your entire campaign. It doesn't mean you can't change it up in different shots. It just means that it has an overall cohesive look and feel. You can use reference images as well and ask Claude or Chat to extracted color palettes, film-grade language, lighting direction, lens choices, composition from them to create your locked style and to help in writing your prompts. I attached a PDF here and in the classroom section for you to download and keep when working on your own images when you want to create a similar color grade or lighting style. I will use my Americana project as an example. Core visual identity Sun-struck 1970s roadside Americana: motel-pool glamour, chrome diner nostalgia, tactile beauty editorial, and intimate direct-gaze portraiture. The look is richly warm and contrasty, not faded or pastel, with real skin texture, sun-baked shadows, restrained teal/cyan environments, and cherry-pink/red fashion accents.

Tonight I put two platforms head to head for talking head video: HeyGen's avatar versus Seedance 2.5 inside Higgsfield. I wanted to know which one felt more real, more natural. Which one actually looks like a person talking ? HeyGen's avatar, or Seedance 2.5 inside Higgsfield? Either? Which do you prefer? Thoughts? The first video: HeyGen, the second video: Seedance 2.5 HeyGen's video turned out to be 22 seconds long, and it's 1920 at 1080 Seedance 2.5 inside of Higgsfield used the full 30 seconds, and it's 1080 by 720 (the largest size available right now in Higgsfield) Here was the script I feed it. I only did one take each. "The thing people get wrong is — okay no, let me back up. (restarts) A camera doesn't take the picture. You take it. The camera just records the decision. (gestures) Same thing here. Same exact thing. The machine isn't deciding anything, it's the fastest hand I've ever worked with and it has zero taste. (beat) So that's the job now. Taste. Honestly it always was — we just used to hide it behind the labor." "According to Claude" - HeyGen bills Avatar IV and V at 20 credits a minute, so a 30-second clip runs 10 credits. Seedance 2.5 inside Higgsfield cost me 195 for the same 30 seconds. One take each, no re-rolls on either. The credits are worth the same money - both platforms are $49 for 1,000 - so that's roughly 49 cents versus $9.55. Call it 20x.

Next Thursday, we'll do a quick virtual coffee Q&A time. I'd love to meet you, stop by, say hi, and ask any questions. It'll be fun!!

It started with a feeling: 🌊 calm, coastal, the sound of waves, muted coastal color. Inside the core foundation section of the classroom you will find the project. Here is a [link] Go check it out, then build your own together here. 👇

I was asked by Luma Labs to create my own skill on their platform. So I thought, okay, what could be useful? One thing I that in my photography that I almost always had to create for all my fashion clients, was to photograph their current collection on a perfectly lit white backdrop. Whether it's a high-end brand like Valentino, or a fashion brand with a Shopify store, most brands at some point need e-commerce and look-book imagery, to clearly properly represent and show the clothing to sell. So that's what I set out to do: make a skill that would do just that, but do it at a very high standard, elevated way with beautiful lighting and bake in an aesthetic that would reference the high-fashion model feel. Here is the video I made for Luma Labs. If you have an account already, you can try out my Luma Skill for free: https://lumalabs.ai/jodijones-skill All of these images below were just either a sketch or a garment hung on a hanger, piece of clothing on the bed or flatlay garment or a snapshot that I put through the skill, and this is a small sample of what came out...