No description
  • Kotlin 96.4%
  • GLSL 2.9%
  • Shell 0.7%
Find a file
Ole-Morten Duesund 6ed3e8e7b5 Propagate camera binding errors to UI
Add an error StateFlow to CameraManager so camera binding failures
are surfaced to the user instead of silently swallowed by
e.printStackTrace(). CameraScreen collects this flow and displays
errors using the existing red overlay UI. Added Log.e with proper
TAG for logcat visibility.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 15:21:38 +01:00
app Propagate camera binding errors to UI 2026-02-27 15:21:38 +01:00
gradle Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
.gitignore Add release signing configuration 2026-01-29 17:23:00 +01:00
build.gradle.kts Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
gradle.properties Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
gradlew Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
gradlew.bat Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
README.md Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
settings.gradle.kts Initial implementation of Tilt-Shift Camera Android app 2026-01-28 15:26:41 +01:00
tiltshift-spec.md Add radial mode, UI controls, front camera, update to API 35 2026-01-29 11:13:31 +01:00

Tilt-Shift Camera

A dedicated Android camera app for tilt-shift photography with real-time preview, touch-based controls, and proper EXIF handling.

Features

  • Real-time tilt-shift effect preview - See the blur effect as you compose your shot
  • Touch-based controls:
    • Single finger drag to move the focus line position
    • Two-finger rotation to adjust blur angle
    • Pinch gesture to adjust blur zone size
    • Pinch in center to zoom camera
  • Zoom controls - Quick presets (0.5x, 1x, 2x, 5x) plus pinch-to-zoom
  • Auto picture orientation detection - Photos saved with correct EXIF orientation
  • GPS location tagging - Optional EXIF GPS data from device location
  • Haptic feedback - Tactile response for all interactions
  • Saves to gallery - Photos saved to Pictures/TiltShift/ folder

Requirements

  • Android 8.0 (API 26) or higher
  • Device with camera
  • OpenGL ES 2.0 support

Building

  1. Open the project in Android Studio
  2. Sync Gradle files
  3. Build and run on a physical device (camera preview won't work on emulator)

Or from command line:

./gradlew assembleDebug

Permissions

  • Camera (required) - For capturing photos
  • Location (optional) - For GPS tagging in EXIF data
  • Vibrate - For haptic feedback

Architecture

The app uses:

  • CameraX - Jetpack camera library for camera preview and capture
  • OpenGL ES 2.0 - Real-time shader-based blur effect
  • Jetpack Compose - Modern declarative UI
  • Kotlin Coroutines & Flow - Asynchronous operations and state management

Project Structure

app/src/main/java/no/naiv/tiltshift/
├── MainActivity.kt           # Entry point with permission handling
├── camera/
│   ├── CameraManager.kt      # CameraX setup and control
│   ├── LensController.kt     # Lens/zoom switching
│   └── ImageCaptureHandler.kt # Photo capture with effect
├── effect/
│   ├── TiltShiftRenderer.kt  # OpenGL renderer
│   ├── TiltShiftShader.kt    # GLSL shader management
│   └── BlurParameters.kt     # Effect state
├── ui/
│   ├── CameraScreen.kt       # Main Compose screen
│   ├── TiltShiftOverlay.kt   # Touch gesture handling & visualization
│   ├── ZoomControl.kt        # Zoom UI component
│   └── LensSwitcher.kt       # Lens selection UI
├── storage/
│   ├── PhotoSaver.kt         # MediaStore integration
│   └── ExifWriter.kt         # EXIF metadata handling
└── util/
    ├── OrientationDetector.kt
    ├── LocationProvider.kt
    └── HapticFeedback.kt

How the Tilt-Shift Effect Works

The tilt-shift effect simulates a selective focus lens that makes scenes appear miniature. The app achieves this through:

  1. Camera Preview → OpenGL SurfaceTexture
  2. Fragment Shader calculates distance from the focus line for each pixel
  3. Gradient blur is applied based on distance - center stays sharp, edges blur
  4. Two-pass Gaussian blur (optimized as separable passes) for quality and performance

License

MIT