The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
NEDETSKIY MIR — 27-Minute Immersive AI Performance
November 2025
A multi-surface projection installation exploring memory, childhood, and the things we leave behind in the dark. Twenty-seven minutes of generative imagery built for a fully enclosed projection hall — walls, floor, and ceiling becoming a single continuous canvas.
The piece moves through chapters: porcelain dolls with cracked faces, a girl marked by streaks of color and grief, ritual spaces stripped of comfort, and a final passage of red water that engulfs the audience entirely.
Role: Director, AI Visual Direction, Pipeline Architecture
Tools: Veo 3 (~85% of all motion content), Kling, Higgsfield, Seedream4, Nano Banana, After Effects
Format: Live multi-surface projection / 27 min / Original score
Premiered: November 2025
Built around the question of what happens when childhood imagery refuses to stay innocent.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Veo 3.1 and Kling 3.0 are both claiming the top spot for native audio and voiceover 🔊 let's see about that..
I picked a challenging character and a scene from a movie most of us know well, and I gave the same prompt to both AI models.
OMG, Why so seriously scary!? 🤡
Which one freaks you out more? 😅🫣 What do you think about the voiceover?
Reference image + Prompt in comments
13 voted
37%
22 voted
63%
35 votes
Closed
prompt:
"Close-up of a character in white face paint and smeared red lipstick, leaning slowly into frame in a dim flickering warehouse, unsettling grin, eyes locked on camera, cinematic grain, shallow depth of field, dark theatrical lighting, 5 seconds, no cuts. Character speaks...
Nice play with light and shadow.
Built “Blue Star” for the Melius Creative Challenge ✨
Blue Star is a surreal animated dream world inspired by Vincent van Gogh paintings and lucid dreaming.
I wanted to create something that felt less like “AI-generated content” and more like drifting through a living painting at midnight.
The workflow was built entirely inside Melius using connected image + video generation nodes to maintain a consistent painterly dream aesthetic across every scene.
The project moves through:
• floating staircases in endless skies
• dreamlike glowing hallways
• trains crossing oceans in the night
• cosmic sunflower fields
• surreal moonlit worlds
The goal wasn’t realism.
Project link: https://app.melius.com/projects/9a45f4e3-69f6-4ef8-b5ed-fc333c2647c8/canvas/fedf41cc-6f43-4843-a766-fc241927b11c
It was emotion, atmosphere, and the feeling of being inside a dream you don’t want to wake up from.
Huge thanks to the Melius team for creating such a visually creative workflow experience 🌌
This is so unique!
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
portfolioreview
The best portfolios tell a story, not just show a grid. Share yours for feedback.
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?