Omnia

AI-Powered Animated Love Story Generator

Engineered a full-stack generative AI pipeline that transforms couples’ narratives into personalized animated short films using multi-modal large language models for story understanding, scene decomposition, and character-consistent image generation.

Implemented a two-stage generation architecture: LLM-driven scene image synthesis with persistent character grounding from reference photos, followed by image-to-video animation with cinematic motion and camera control.

Built a React frontend with conversational story input flow and a Node.js/Express backend orchestrating model inference, video assembly, and cloud storage via Cloudflare R2.

Tech Stack: Python, React, Node.js/Express, Stable Diffusion, AnimateDiff, Cloudflare R2, CUDA