This is a submission for the Google AI Studio Multimodal Challenge
What I Built
I built Travel360, an app that offers guided tours of a city’s top landmarks through Google Maps’ immersive 3D view. Along the journey, users can capture AI-powered selfies at iconic spots.
Travel360 turns repetitive selfies into unique, shareable moments and makes city exploration fun, accessible, and engaging — without the cost or limits of physical travel.
With just a camera or a simple drag & drop, Gemini’s Nano Banana AI instantly places users in front of world-famous monuments like Times Square, Tower Bridge, or the Statue of Liberty.
Demo
Here’s a walkthrough of Travel360 in action:
1) Navigate to a city in immersive view : https://travel360-50134736379.us-west1.run.app/
2) Click on start tour
3) Upload or take a selfie using camera.
4) Generate a realistic selfie in the selected place.
Current deployed version demos (added features : quick capture mode, and selfie custom weather & daytime) : View NEW live demo :
How I Used Google AI Studio
I leveraged Google AI Studio to bring Gemini’s multimodal capabilities into Travel360 and speed up development:
Code & Debugging Support: Gemini helped generate React/TypeScript snippets, fix bugs, and streamline integration with Google Maps immersive 3D view.
Landmark Data: Gemini suggested the best places to feature in each city and provided their geographic coordinates to build a guided tour experience.
Selfie Generation (Prototype): With Gemini Nano Banana, I connected uploaded user images with context prompts (e.g., “Place the user at Times Square”) to generate realistic selfies at landmarks.
Multimodal Flow: Gemini processed both image + text inputs and produced image outputs, enabling a seamless flow between the guided city tour and selfie creation.
This mix of AI-assisted development and multimodal integration made Travel360 both faster to build and more engaging to use.
Multimodal Features
Travel360 enhances the user experience through multimodality in several ways:
Image + Text Inputs → users upload a photo and Gemini interprets instructions (via prompt) about where to place them.
Immersive Map + AI Output → Google Maps 3D provides the navigation, while Gemini (gemini-2.5-flash-image-preview) generates a personalized selfie at the selected landmark.
By combining the Google Maps immersive API with Gemini’s multimodal image generation API, Travel360 delivers a unique and playful experience — making every city feel instantly accessible.
Top comments (16)
Good work ! Amazing
Wonderful 👏
really great app !
Really impressive !
Good luck Amine
Un grand bravo pour cette utile application 👍
Bonne chance et très bonne continuation.
Good luck
Un grand bravo pour la création de cette application très très utile !
Je te souhaite une très bonne continuation et bonne chance.
Goooood work !
I like it ! Very innovative
Some comments may only be visible to logged-in visitors. Sign in to view all comments.