appssemble
ServicesBlogCase StudiesAboutContact
Case Studies/Dezigner

Snap objects, remove backgrounds, place in AR

Photograph any object, remove its background instantly with on-device ML, organize collections into mood boards, and place items in real spaces using AR. No internet needed — all models run locally.

View on App Store
removeobjectCOLLECTION+AR PLACEMENTML STATUS100% offlineFEATURESsnap & removecollectmood boardsAR placeexport PNG
100%offline capable
5.0App Store rating
On-devicebackground removal
What it does

Capture, collect, place

01

Snap & Remove Background

Take a photo or pick from your gallery. The on-device ML model removes the background instantly — no internet required. Works on furniture, art, plants, anything. Clean cutouts every time.

Background removalOn-device MLOffline
02

Object Collection

Save every cutout to your personal collection. Organize items into themed groups. Build a visual library of objects you can reuse across mood boards and AR scenes.

CollectionsOrganizationVisual library
03

Mood Boards

Arrange collected objects into mood boards. Change background colors to test combinations. See which pieces work together before committing to a design. Export boards as images.

CompositionColor testingExport
04

Augmented Reality

Place any collected object in real space using AR. See how furniture fits in a room, how art looks on a wall, or how a collection works in context. No internet needed — all rendering happens on-device.

ARKitReal-space previewOn-device rendering
05

Edit & Export

Swap backgrounds, adjust sizing, and fine-tune cutouts. Export clean PNGs for use in other design tools. The app fits into existing creative workflows without replacing them.

PNG exportBackground swapDesign tool integration
Under the hood

What powers it

Swift

Native iOS app optimized for camera, ML inference, and AR rendering.

TensorFlow Lite

Custom on-device ML models for real-time background removal without internet.

ARKit

Apple's AR framework for placing collected objects in real-world spaces.

Core ML

Apple's ML framework for optimized on-device inference and model management.

Core Image

Image processing pipeline for cutout refinement and background manipulation.

Figma

Design system created for creative professionals and interior designers.

Process
Agile2-week sprintsDaily standups
Technologies
iOS (Swift)TensorFlowARKitFigma
Takeaways

What we learned

01

Offline-first is hard, but worth it

Making every feature work without internet was the biggest challenge. In-house ML models for background removal had to be small enough for mobile yet accurate enough for professional use. The constraint produced better engineering.

02

Tech meets creativity

Building an app for creative professionals pushed our designers and engineers to think differently. Our designers built for people like themselves. Our engineers loved the puzzle of real-time ML on mobile.

03

ML opens new doors

Working with custom machine learning models that identify patterns and improve over time showed us how vast the possibilities are. Every project since benefits from what we learned building Dezigner.

Case studies

More work

10M+events / day
Grovs
Open SourceAttributionDeep Linking
→
500K+app sessions
Semaphr
SaaSiOS SDKAndroid SDK
→
<30sdraft to e-Factura
Incasez
Invoicinge-FacturaSaaS
→
8 RAWframes merged
HDR Plus+
iOSMLTensorFlow
→
Let's talk about your project
[email protected]
CompanyServicesCase StudiesBlogContact
Offices
New York1740 Broadway, 15th Floor, 10019
LondonKemp House, 160 City Road, EC1V 2NX
Cluj-NapocaBlvd. 21 Decembrie 1989, 95-97
SocialLinkedInGitHub
© 2026 appssemble. All rights reserved.
Privacy PolicyCookie PolicyJobsGlossary