Perceive - Point. Ask. Learn. - AI insights at your fingertips

Perceive is a fully AI powered Android App that helps you to explore, learn and play like never before! Just point your camera at anything and start a conversation! See something interesting and want to know more about it? Just point the camera and ask away!

Table of Contents

1. Demo
2. Screenshots
3. Tech Stack
4. Notable Features
5. Source code, Architecture, & Testing
6. Building and running the app

Demo

Coming Soon!

Screenshots

Tech Stack

• Entirely written in Kotlin.
• CameraX for accessing and processing images captured from the camera.
• Hilt for dependency injection.
• Jetpack Compose for UI and navigation.
• Preferences Datastore for persistently managing key-value pairs.
• Kotlin Coroutines for threading.
• Kotlin Flows for creating reactive streams.
• Kotlin Gradle DSL and Gradle Version Catalogs for managing dependencies.
• Gemini Client SDK for using Google's Gemini-pro models.

Notable features

Haptic Feedback 🪄

The app subtly improves the user experience by leveraging the on-device vibration actuators to provide haptic feedbacks when certain events occur in the app such as muting the assistant and tapping the mic.

Dynamic Colors & Light/Dark Themes 🎨

The app seamlessly adapts to your device's theme preferences. It utilizes dynamic colors when available and offers beautiful, pre-defined light and dark themes for a consistent and enjoyable experience.

Themed App Icons (Android 13+) 🌈

The app also supports the "Themed Icons" feature available on Android 13 and later. If the user has opted in for the feature on a device that is running Android 13 and higher, the app's icon will be tinted to inherit the coloring of the user’s chosen wallpaper and other themes.

Intelligent Cache Management ✨

The app automatically clears cache data once it's no longer required, ensuring efficient resource utilization.

One time welcome screen 🙋🏼

The app greets users with a captivating welcome screen (displayed only on the first time the user launches the app after installation), ensuring a simple but effective onboarding experience.

Source code, Architecture, & Testing

• Each package is considered as a separate cohesive module/sub-module representing a layer or a feature.
• The top level packages are divided by layers, and each sub-package within them represent a single individual feature.
• Commit messages follow the Conventional Commits specification.

Building and running the app

1. Generate a new API key to access the Gemini API from here.
2. In the local.properties file of your project add the api key as shown below.

GOOGLE_GEMINI_API_KEY = PASTE-YOUR-API-KEY-HERE

3. Build the app and run it.