I built Earth Zoom Out, a web app that takes any uploaded image and transforms it into a stunning 6-second video that zooms out to reveal Earth from space.
Key features:
• Upload any image and watch it transform into a cinematic zoom-out sequence
• Built with Next.js and powered by AI video generation
• User authentication and video history
• 6-second output videos perfect for social media
The idea came from wanting to create that dramatic "cosmic perspective" effect you see in movies. Users can upload a photo of their house, vacation spot, or any location, and see it
contextually placed within our planet.
Live demo: https://earthzoomout.co/
Tech stack: Next.js, TypeScript, Tailwind CSS, NextAuth, Fal.ai for video generation.
I'd love to hear your thoughts and feedback!
One of the biggest challenges in AI art is keeping the same character recognizable across different scenes. This tool tackles that problem in a few clever ways: - Action presets – instead of rewriting prompts, you can pick from predefined poses and actions. - Character management – save and reuse characters, so you don’t have to rebuild them from scratch each time. - Flexible styles – works well across anime, semi-realistic, and other common art styles.
I’ve been testing it, and while it’s not perfect (character drift is still a thing), it makes the process much smoother compared to traditional prompt-only methods.
Here’s the link if you want to try it out: https://aicharactergenerator.app/
Curious if anyone else has tried similar approaches to solving character consistency in AI art?