Liudmyla Biliaieva
Offer Liudmyla work on your next project.
Private message
Offer work
Kagarlyk,
Ukraine
3 months 21 days back
available for hire
on the service 4 months 15 days
Rating
Successful projects
Average rating
Language proficiency level
Українська: fluent
Русский: fluent
English: elementary
Skills and abilities
Portfolio
-
20 USD AI-based visual for a premium fitness instructor brand
Photo Processing
Goal: Create a visual ecosystem for the social media of a rehabilitation specialist. The main task is to provide regular high-quality content without the need for weekly expensive photo shoots.
Client's problem: The instructor needed diverse content (from business portraits to working moments in the gym), but a tight training schedule and the high cost of renting premium locations hindered the creation of professional visuals.
… My solution:
Developed a unique digital model based on real photographs of the client.
Created a series of photorealistic images in various settings: from formal (Holiday/Formal) to functional (Luxury Gym).
Ensured perfect facial resemblance (Face Consistency) and athletic physique across all shots.
What I achieved (Impact/Results)
Budget savings: Reduced content costs by 80% due to the absence of expenses for a photographer, studio, and stylist.
Visual unity: Created a cohesive profile style that enhances trust among a premium audience.
Content readiness: Established an image database for 3 months ahead in just one working week.
-
Cinematic AI video, maintaining character consistency
Video Creation by Artificial Intelligence
"Breaking Reality" — Cinematic AI Video
This is my final project for the AI Content Maker course. The main goal: to implement the concept of "Breaking Reality" — transforming a real person into various stylized universes while maintaining 100% character recognition (Character Consistency).
Technical Tasks and Solutions
… Preserving Identity: The main challenge was to retain the hero's facial features when transitioning from the real world to styles such as Cyberpunk, Medieval Fantasy, and 3D animation. This task was accomplished using advanced prompt engineering and style transfer methods.
Dynamic Production: Combining motion design and AI-generated video to create seamless transitions between "realities."
Visual Styles: Three aesthetics were realized: High-end Cyberpunk, Cinematic Medieval, and stylized 3D graphics (in the style of Pixar).
Technology Stack
Base Generation (images): Midjourney, Krea.
Video and Animation: Kling.
Post-production and Editing: CapCut.
Psychological Aspect
As a professional psychologist, I use AI as a tool to explore human archetypes. This project illustrates the "Multiverse of Personality" — the idea that identity remains recognizable and authentic, regardless of visual or narrative context.
Please watch the video in the portfolio
https://www.behance.net/gallery/243239243/AI-Multiverse-Character-Consistency