Android App - Offline AI Background Eraser & Ad Creator

ANDROID DEVELOPER NEEDED – OFFLINE AI BACKGROUND REMOVER & AD MAKER

PROJECT OVERVIEW

I need an experienced Flutter/Android developer to build a production-ready Android application combining an offline AI Background Remover with a Product Photo and Micro-Ad Studio.

The app workflow:

SELECT PHOTO → REMOVE BACKGROUND → EDIT → CREATE AD → EXPORT

The most important requirement is that the core AI background removal and image editing work directly on the Android device without a paid cloud AI/API.

PREFERRED TECHNOLOGY

• Flutter + Dart preferred
• Native Android/Kotlin also accepted
• TensorFlow Lite, ONNX Runtime, MediaPipe or another suitable on-device ML solution

1. OFFLINE AI BACKGROUND REMOVAL

The background-removal AI must run locally on the device.

NO Remove.bg API or other paid cloud background-removal API.

The developer must propose the exact AI model before development and provide:

• Model name
• Model size
• License
• Commercial-use/redistribution rights
• Expected performance
• Android compatibility
• RAM requirements

The model must legally allow commercial use and redistribution inside the app.

The user photo must not be uploaded to an undisclosed server.

Core workflow must work offline:

PHOTO → AI BACKGROUND REMOVAL → EDIT → EXPORT

Internet may only be required for services such as AdMob, Google Play Billing, Play Store updates or other clearly disclosed services.

2. IMAGE IMPORT

Users should be able to:

• Select photos from Gallery
• Capture photos using Camera
• Start a new project
• Open saved projects

The app should handle normal and high-resolution images safely.

3. AI PROCESSING & MANUAL EDITING

After selecting an image:

• Process locally using the AI model
• Detect the foreground
• Remove the background
• Generate transparent PNG/cutout
• Show preview

Manual correction tools:

• Eraser brush
• Restore brush
• Brush-size control
• Zoom
• Pan
• Undo
• Redo

The app should be optimized for normal and mid-range Android phones.

Target AI processing time is approximately 2 seconds for typical images where device hardware allows. This is a performance target.

4. MAIN CANVAS STUDIO

The user can create promotional designs from the cutout.

Canvas presets:

• 1:1
• 9:16
• 4:3
• Custom size if practical

Cutout controls:

• Move
• Scale
• Rotate
• Flip
• Position

The editor should support multiple layers.

5. BACKGROUNDS

Solid colors:

• White
• Black
• Custom color

Gradients:

• Linear
• Radial
• Custom colors

Bundled offline backgrounds:

• Marble
• Wood
• Concrete
• Studio backgrounds
• Simple textures

All built-in backgrounds must work offline.

6. DROP SHADOW / GLOW

Provide controls for:

• Shadow opacity
• Blur
• X/Y offset
• Shadow color
• Optional glow

7. TEXT EDITOR

Users can add multiple text layers.

Features:

• Font selection
• Font size
• Color
• Bold/normal
• Alignment
• Letter spacing
• Line spacing
• Background box
• Stroke/outline
• Shadow
• Move
• Scale
• Rotate

Include approximately 10–15 properly licensed fonts.

8. E-COMMERCE STICKERS

Include a local sticker/badge library.

Examples:

• 50% OFF
• SALE
• NEW
• BEST SELLER
• FREE SHIPPING
• LIMITED OFFER
• NEW ARRIVAL
• PRICE TAG
• Star ratings

All assets must be commercially licensed or created for this project.

9. EXPORT

Export locally as:

• PNG
• JPEG/JPG

PNG should support transparency.

Images must save to the Android Gallery using appropriate modern Android storage/MediaStore methods.

10. LOCAL PROJECT SAVING

Users must be able to save unfinished designs and reopen them later.

Projects should preserve:

• Original image
• Cutout
• Background
• Text
• Stickers
• Positions
• Scale
• Rotation
• Shadows
• Canvas dimensions
• Other editing settings

No cloud account should be required for normal editing/project storage.

11. GOOGLE ADMOB

Integrate Google AdMob according to current policies.

Banner:
Display appropriately without blocking editing controls.

Interstitial:
Can be shown after suitable actions such as exporting. Do not show excessive/disruptive ads.

Rewarded Ad:
Can unlock selected premium features such as:

• Premium template
• Premium sticker
• Batch processing
• Other temporary Pro features

Rewards must only be granted after a valid completed rewarded advertisement.

12. PRO / IN-APP PURCHASE

Implement a one-time Google Play Pro purchase.

Possible benefits:

• Remove advertisements
• Pro templates
• Pro stickers
• Additional backgrounds
• Other Pro features

Purchase handling must support purchase state, restoration/reinstallation and failed/cancelled transactions.

13. UI/UX

The app should look modern, clean, professional and easy for beginners.

Main workflow:

SELECT PHOTO → REMOVE BG → EDIT → CREATE AD → EXPORT

Main areas:

• Home/Image Picker
• AI Processing
• Manual Touch-Up
• Main Studio
• Backgrounds
• Text
• Stickers/Templates
• Export
• Settings/Pro/About

14. TECHNICAL REQUIREMENTS

Preferred architecture:

Flutter + Dart

Possible components:

• Flutter UI
• Native Android integration where required
• TFLite/ONNX/local ML
• Local database/storage
• Android MediaStore
• Google Mobile Ads SDK
• Google Play Billing

Optimize for:

• RAM usage
• Large images
• ML model loading
• Canvas performance
• Export speed
• App startup
• Mid-range devices

15. SOURCE CODE & OWNERSHIP

Complete source code and project ownership must be transferred to me after final payment.

Required delivery:

• Complete source code
• Git repository
• Flutter/Kotlin project
• ML model files
• ML implementation
• Configuration files
• Android project
• App assets
• Fonts and licenses
• Stickers/SVG files
• Documentation
• APK
• Production AAB

No essential functionality may depend on the developer's personal server, account or undisclosed API.

All third-party libraries, SDKs, AI models, fonts, assets and services must be disclosed.

16. NOT ALLOWED

• Paid cloud AI for background removal
• Remove.bg API
• Undisclosed backend/server
• Uploading user photos without disclosure
• Pirated/unlicensed assets
• AI models without commercial redistribution rights
• Undisclosed tracking/analytics
• Hidden recurring costs
• Source-code lock-in
• Dependence on developer's personal accounts

17. TESTING

The developer must test:

• Offline background removal
• Large images
• Mid-range Android devices
• Manual eraser/restore
• Canvas editing
• Text
• Stickers
• Shadows
• PNG/JPEG export
• Gallery saving
• Project reopening
• AdMob
• Rewarded ads
• IAP
• Memory/crash behavior

18. MILESTONES

Milestone 1 – AI Prototype
• Project setup
• Image import
• Offline AI model
• Working background removal

Milestone 2 – Editor
• Manual editing
• Canvas
• Cutout controls
• Backgrounds
• Shadows

Milestone 3 – Ad Studio
• Text
• Stickers
• Templates
• Export
• Local projects

Milestone 4 – Monetization
• AdMob
• Rewarded ads
• Google Play Billing
• Pro unlock

Milestone 5 – Testing
• Device testing
• Optimization
• Bug fixing
• Offline testing

Milestone 6 – Final Handover
• Source code
• APK
• AAB
• Documentation
• Assets

19. PROPOSAL REQUIREMENTS

Please answer:

1. Flutter or Native Kotlin? Why?
2. Which exact offline AI model will you use?
3. What is its commercial license?
4. Have you implemented on-device image segmentation before?
5. Show similar projects.
6. How will you prevent Out Of Memory errors?
7. How will you implement the interactive canvas?
8. How will projects be stored locally?
9. Have you integrated AdMob?
10. Have you implemented Google Play Billing?
11. What Android versions will you support?
12. What devices will you test?
13. Estimated development time?
14. Fixed total price?
15. Post-delivery bug-fix/warranty period?

Please do not send a generic proposal. I am specifically looking for a developer experienced in Flutter/Android, on-device AI, image segmentation, image processing and interactive canvas development.

FINAL REQUIREMENT:

This must be a real production Android application, not a UI prototype.

BACKGROUND REMOVAL AND CORE EDITING MUST WORK LOCALLY ON THE ANDROID DEVICE WITHOUT A PAID CLOUD AI/API.

Complete source code, assets, ML model/license information and project ownership must be delivered at the end of the project.