IMIP Artefact Demo - 8 May 2026

Arclight

An AI sticker archive for everyday objects.

Artefact + Report15-minute demoiOS-first app
From one object photo to a small visual memory.
BFL-generated sticker style contact sheet
Creative Memory

Turn objects into memories.

Arclight makes a sticker, adds metadata, and lets me edit the record.

A sticker becomes a memory when it keeps the object's context.

Creative memory means keeping a small everyday object as a visual marker with useful information.

Arclight can generate and restyle a sticker. It also saves editable metadata, so the object is easier to find later.

01
AI draft

AI suggests the object name and category.

02
User edit

The user can correct or add details.

03
Memory record

The sticker keeps useful context, not only a nice look.

Prototype Evidence

Prototype evidence

31 V1 and V2 screens. Click a section for details.

Click a surface for iteration details.
V1 Homepage 01.1 V1 Homepage 01.1
V1 Homepage 01.2 V1 Homepage 01.2
V1 Homepage 02 V1 Homepage 02
V1 Homepage 03.1 V1 Homepage 03.1
V1 Homepage 03.2 V1 Homepage 03.2
V1 Homepage 04.1 V1 Homepage 04.1
V1 Homepage 04.2 V1 Homepage 04.2
V1 Homepage 05.1 V1 Homepage 05.1
V1 Homepage 05.2 V1 Homepage 05.2
V2 Homepage 01 V2 Homepage 01
V2 Homepage 02 V2 Homepage 02
V1 Library 01 V1 Library 01
V1 Library 02 V1 Library 02
V2 Capture 01 V2 Capture 01
V2 Capture 02 V2 Capture 02
V2 Capture 03 V2 Capture 03
V2 Review 01.1 V2 Review 01.1
V2 Review 01.2 V2 Review 01.2
V2 Review 01.3 V2 Review 01.3
V2 Review 02 V2 Review 02
V2 Review 03 V2 Review 03
V2 Review 04 V2 Review 04
V2 Review 05 V2 Review 05
V2 Review 06.1 V2 Review 06.1
V2 Review 06.2 V2 Review 06.2
V2 Detail 01.1 V2 Detail 01.1
V2 Detail 01.2 V2 Detail 01.2
V2 Detail 02.1 V2 Detail 02.1
V2 Detail 02.2 V2 Detail 02.2
V2 Calendar 01 V2 Calendar 01
V2 Calendar 02 V2 Calendar 02
Optional Prototype Inspection

Home Iteration

Home changed from an intro screen into a sticker board.

Home changed from an intro screen into a sticker board.

  • V1 explained the idea and pushed camera access.
  • V2 shows stickers and object labels first.
  • Home becomes the archive surface, not only navigation.
V1 Homepage 01.1
V1 Homepage 01.2
V1 Homepage 02
V1 Homepage 03.1
V1 Homepage 03.2
V1 Homepage 04.1
V1 Homepage 04.2
V1 Homepage 05.1
V1 Homepage 05.2
V2 Homepage 01
V2 Homepage 02
Optional Prototype Inspection

Library Iteration

The separate Library page moved into Home.

The separate Library page moved into Home.

  • V1 explored categories and saved items.
  • V2 removes the extra Library layer.
  • Home becomes both display and archive entry.
V1 Library 01
V1 Library 02
Optional Prototype Inspection

Capture Iteration

Capture became simpler and more focused on real objects.

Capture became simpler and more focused on real objects.

  • Early capture screens had extra options.
  • V2 makes the camera button the main action.
  • The simpler flow supports real object capture.
V2 Capture 01
V2 Capture 02
V2 Capture 03
Optional Prototype Inspection

Review Iteration

Review is where the user checks AI output before saving.

Review is where the user checks AI output before saving.

  • Cutout and recognition are visible before saving.
  • The user chooses the style.
  • AI output becomes an editable decision.
V2 Review 01.1
V2 Review 01.2
V2 Review 01.3
V2 Review 02
V2 Review 03
V2 Review 04
V2 Review 05
V2 Review 06.1
V2 Review 06.2
Optional Prototype Inspection

Detail Iteration

Detail shows metadata, not only the generated image.

Detail shows metadata, not only the generated image.

  • AI suggests a name and category.
  • The user can correct metadata after generation.
  • Location, notes, and category help people revisit the sticker.
V2 Detail 01.1
V2 Detail 01.2
V2 Detail 02.1
V2 Detail 02.2
Optional Prototype Inspection

Calendar Iteration

Calendar makes saved stickers visible by date.

Calendar makes saved stickers visible by date.

  • Saved stickers become memory markers.
  • The archive can be revisited through dates.
  • This supports the creative-memory idea.
V2 Calendar 01
V2 Calendar 02
Iteration

From camera-first to object-first.

I reduced navigation and made saved objects more visible.

Home From

Camera entry and explanation

To

Sticker board
Home now shows saved stickers and objects first.

Library From

Separate category page

To

Part of Home
This removes one extra navigation layer.

Capture From

Import, flash, options

To

One main camera action
This pushes the demo toward real objects.

Project Aims

My goal: an editable object memory.

The app captures, understands, restyles, edits, and saves one object.

01Capture a real object
02Cut it out from the photo
03Recognise it and draft metadata
04Let the user edit metadata
05Apply the chosen art style
06Save it as a local memory record
System Pipeline

The app works as a pipeline.

Each step turns a real object into a saved record.

01
Camera
The user photographs a real object.
02
iOS Vision cutout
The iPhone extracts the foreground object locally.
03
Gemini metadata
AI suggests name, category, and context.
04
Review / edit
The user checks the cutout and edits metadata.
05
Style choice
The user picks the art direction.
06
FLUX Worker
Cloudflare Worker creates the styled sticker.
07
Local archive
SQLite and FileSystem save the editable record.
Design Decisions

AI should stay editable.

The user checks metadata, chooses style, and decides what to save.

Object-first Home

Home shows stickers and saved objects first.

Library inside Home

The archive is easier to find.

Simple Capture

One clear camera action keeps the demo focused.

Review before save

AI gives a draft; the user checks it.

Editable local archive

Saved objects can be reopened and changed.

Style System

Why these four styles?

Hockney, Kusama, Matisse, and Monet show wider contrast across modern and classical art.

Old set Custom pencil, Degas, Renoir, Monet

These options felt too close: soft drawing, painterly colour, and impressionist mood.

New set More art languages

The new set lets one object move between modern colour, pattern, cut-paper shape, and classical light.

Hockney style sticker reference
HockneyFlat colour / modern space
Kusama style sticker reference
KusamaDots / repetition / pattern
Matisse style sticker reference
MatisseCut paper / bold shapes
Monet style sticker reference
MonetLight / classical impressionism
Live Demo Map

Now I will show the live loop.

Capture an object, check it, style it, save it, and revisit it.

Demo flow 1: Home 01 Home
Demo flow 2: Camera 02 Camera
Demo flow 3: Photograph 03 Photograph
Demo flow 4: Review 04 Review
Demo flow 5: Metadata 05 Metadata
Demo flow 6: Style 06 Style
Demo flow 7: Save 07 Save
Demo flow 8: Board 08 Board
Demo flow 9: Calendar 09 Calendar
Demo flow 10: Detail edit 10 Detail edit
Switching To Phone

Capture. Check. Revisit.

I will keep this slide up during the phone demo.

Checkpoint 01Capture

Photograph a real object.

Checkpoint 02Check

Check cutout, metadata, and style.

Checkpoint 03Revisit

Save, then open Home, Calendar, and Detail.

Implementation Evidence

Built as an iOS-first prototype.

Native cutout, Cloudflare AI Workers, local storage, and tests support the demo.

App

Expo 54, React Native 0.81, Expo Router 6

Native

Local Expo module, iOS Vision subject cutout

AI services

Cloudflare Workers for Gemini and FLUX

Storage

Expo SQLite records and FileSystem images

Verified

npm run lint passed.

npx vitest run passed with 21 test files / 84 tests.

These checks support the prototype demo. They are not user research.

Limits

Main limits.

The prototype works, but it depends on iOS, network AI calls, and metadata review.

iOS build

The full pipeline needs the iOS 17 development build.

Worker setup

Recognition and style generation need valid Worker URLs and API keys.

Network

AI calls need internet, so I keep a backup recording.

Metadata errors

AI can be wrong, so Review/Edit is necessary.

Evaluation

The evidence is smoke tests and automated tests, not a broad user study.

Future Work

Next work.

Test with users, add more styles, diversify backend services and AI models, and improve editing.

01User testing

Test whether people use sticker memories to revisit real objects.

02More styles

Support more art styles, backend services, and AI models.

03Control

Improve metadata editing, style consistency, and offline fallback.

Closing Claim

Arclight makes editable sticker memories.

The app connects capture, AI style, metadata, and a local archive.

BFL-generated sticker style contact sheet
Questions

Any questions?

01 / 15
Arclight - IMIP Demo