The Vulture — "Temptation" (Pt. IV)

Inspiration

This project is inspired by my original sci-fi novel series 'The Elevator'.

What it does

This is the fourth and final, 10min epic in 'The Vulture' series — a cinematic, phantasmagoric collection of vignettes exploring the subconscious lives of four women whose pasts, traumas, and destinies converge in the apocalyptic world of my novel series, 'The Elevator'.

Set within a symbolic dreamworld, each short captures a woman’s private encounter with Death. In Sheena’s nightmare world, Death tempts and lures. In Monica’s world, it lurks and feeds. In Jilisa’s, it isolates and taunts. Zoey’s world — she welcomes it.

Each vignette is one part memory, one part nightmare, one part prophecy of what’s to come.

The videos were designed as a means to promote my novel through captivating, cinematic storytelling. I approached this concept not just from a video creator's perspective but also as a writer. I tapped into the symbolic essence of what each character endures before the core plot of my novel series and then I created a visual nightmare. It was a great way to showcase something gripping about the character while also keeping audiences guessing — no spoilers. Sheena is falling deeper and deeper into hopelessness, unaware just how close she is to Death who's made her its next prey.

This concept captures the dueling sides of a bright spirit battling the dark scourge of trauma and rejection.

How I built it

I used LTX Studio to design my principal characters featured throughout the series, including the recurring character of Death. This was my most ambitious project yet. Envato has been hosting an unlimited generations promotion on their site. So, I used its image generator to create the backgrounds for each shot, then layered cutouts of my characters with the backgrounds in Pixelmator Pro to create the scene. For the more complicated shots involving multiple characters and supernatural elements I did a mix of this workflow—designing in Pixlemator Pro, tweaking poses and expressions with the image generator, rinse and repeat until I have my start frame. I ventured into experiments with continuous shots when Envato added the End Frame feature to its platform in the middle of building this project. Whenever the system altered my character's facial features I would either tweak the start frame image or convert the video result using FaceTool. Swapping the wrong face in the video with the headshot I generated with LTX. Admittedly, I took total advantage of the promotion and have an overabundance of generations, choosing to be less forgiving with the results until I achieved the shot I needed/wanted. The workflow involved: • Character masking • Building scene backgrounds in Envato with detailed prompts • Layering characters in each scene either through Envato's Nano Banana feature or Pixlemator Pro, or both • Color grading, photo enhancements/repair (Nano Banana degrades the image after multiple rounds of layering) • Video generations: Adding a start frame and writing a detailed prompt in the following sequence — Camera angle/motion/quality, Character description, Character emotion & action, any additional scene/character action, scene tone/mood. • Final Cut Pro/Apple Motion: Editing all acquired video assets into sequences and adding motion graphics for the title graphic and end screen. • Voice Over: Recording a simple voice memo and converting the voice with a Custom Voice in ElevenLabs using Speech To Text.

Challenges we ran into

Because of the complexity of my project (and personal deadlines) it was hard to achieve total character consistency. But, honestly, I like leaving traces that these people are fake. The realism is incredible but I don't wish to trick anyone. I just love being able to create what I see in my head. Without, the kind of budget it takes to pull off a vision like this with live casts and crew, it would have just continued to live on quietly in my brain. Having multiple prominent characters in a shot was definitely the biggest achievement in this whole project, and of course, the most difficult. Generating each individual character image, masking, then layering them together. Making them all move with consistent facial identity, maintaining their costumes, operating their props realistically...this was all very difficult to achieve. I had to adjust my camera angles—thusly, rebuilding the shot. But, eventually I achieved a scene I'm very happy with.

Accomplishments that I'm proud of

The video as a whole is a source of pride. But as mentioned above, achieving scenes with a full cast, as well as some pretty complex supernatural concepts—these were not easy. So, super proud of how it all worked out. The promotion really gave me the time and ability to learn how to effectively build a scene and prompt it into moving picture. When I started editing with AI, I prompted and just took whatever was usable. I'd come to expect that whatever I prompted wasn't going to actually work. Now, I experience—more times than not—the exact visual I ask for...the tricky part is working around character inconsistency and AI glitches in the results.

What I learned

1. With time and patience (which I don't have), you can do almost anything. AI is truly a power to wield honorably and with great respect.
2. AI is constantly evolving at an incredible rate. It's never the same by the time I've finished a project.
3. Understanding the various models and what their strengths are.

What's next for The Vulture — "Temptation" (Pt. IV)

This is the end of The Vulture series...for now.

Built With

• applemotion
• elevenlabs
• envato
• finalcutpro
• ltx
• nanobanana
• pixelmator