Skip to content

CivicWear VTO - E #16

@Chai9anya

Description

@Chai9anya

Team Name

Code Blooded

Hackathon Track

App/Dev

ps-number

1

Email Address

patil979chaitanya@gmail.com

Email Addresses of Team Members

naaz.shaikh@vit.edu.in, fatima.shaikh@vit.edu.in, samarth.dayma@vit.edu.in

Project Description

Civicwear is a premium AI-powered fashion atelier that redefines the virtual try-on experience through its advanced VTO-E synthesis engine. It allows users to realistically visualize both curated catalog items and their own custom garment uploads on a personalized profile. With a dedicated Designer Studio for bulk collection management and a sophisticated editorial aesthetic, Civicwear bridges the gap between personal creativity and high-end digital retail.

Inspiration behind the Project

Civicwear is a Digital Atelier designed to bridge the gap between flat design and human expression. It transforms the traditional e-commerce experience into a high-fidelity, interactive studio where the user is both the model and the curator.

Tech Stack

Core Frontend
React 19: The latest functional component architecture for a reactive UI.
TypeScript: Ensuring type-safe data structures for products and user profiles.
Vite 6: A lightning-fast build tool and development server.
Styling & Motion
Tailwind CSS 4: Utility-first styling for the premium "Brand Gold" and "Espresso" aesthetic.
Motion (Framer Motion): Powering the immersive transitions, staggered reveals, and the VTO-E modal effects.
Lucide React: A clean, consistent iconography set for the studio and catalog.
AI & Synthesis Engine
Google Gemini API (@google/genai): The core of the VTO-E Synthesis Engine.
Model: gemini-2.5-flash-image — specifically tuned for high-fidelity localized garment synthesis and anatomical constraint locking.
Data & Assets
Browser LocalStorage: Providing persistent storage for your Designer Studio collections across sessions.
FileReader API: Handling high-resolution bulk image uploads and base64 encoding for AI processing.
Anatomical Constraint Prompting: A custom-engineered system instruction set that ensures identity preservation and skeletal ratio locking during the try-on process.

Project Repo

https://ai.studio/apps/ee210b29-e57f-43c8-bef8-fb28e3f4f898

Demo Video

https://drive.google.com/file/d/19ZflytxeOJxtUvWbMyjCtiCmoiWabiOU/view?usp=drivesdk

Presentation Link

https://docs.google.com/presentation/d/1LSEW_WPMYTpD0LRuAtmL9e4Unyy9LqXk/edit?usp=drivesdk&ouid=103151546261512433999&rtpof=true&sd=true

Anything Else?

No response

Rules and Code of Conduct

  • I agree to follow this hackathon's Rules and Code of Conduct

Metadata

Metadata

Assignees

No one assigned

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions