DreamLensAI

Turn children’s words and dreams into magical images and videos.

Comment

Inspiration

People often imagine stories, dreams, and creative ideas but lack simple tools to turn them into visuals. Existing AI tools can be complex or require technical knowledge. This inspired us to create DreamLensAI, a platform that helps anyone easily transform their ideas into images and videos.

What it does

DreamLensAI allows users to type or speak their ideas and instantly generate AI-powered images or animated videos. It transforms imagination into visual content, helping users go from “I imagined this” to “I can see it.”

How we built it

We built the frontend using React and the backend using FastAPI.
We used Amazon Nova Pro for prompt enhancement, Nova Canvas for image generation, and Nova Reel for video generation. Media files are stored in AWS S3, while PostgreSQL stores user data and media URLs.
The application is containerized using Docker for both frontend and backend and is deployed on Render.

Challenges we ran into

Integrating multiple AI models into separate image and video generation pipelines was challenging. We also needed an efficient way to store large media files, which we solved using AWS S3 while keeping only media URLs in the database.

Accomplishments that we're proud of

We successfully built a working multimodal AI platform that converts human ideas into images and videos. The system integrates multiple AI models while providing a simple and accessible creative experience for users.

What we learned

We learned how to design and integrate multimodal AI pipelines, manage cloud storage for large media assets, and connect a full-stack application using React, FastAPI, Docker, and cloud deployment platforms.

What's next for DreamLensAI

Next, we plan to add voice-narrated stories, AI-generated storybooks, character builders, multilingual support, and more creative tools to make DreamLensAI a powerful storytelling and creativity platform for everyone.

Built With

Share this project:

Updates