Let the object lead
Use cutout images, visual ordering, grid and shelf views so collectors can browse the collection visually.
↳ Documented product behavior← Selected work
Mobile product · Computer vision / 2026
CanDex
A visual catalog for can collectors, combining camera capture, background removal, barcode recognition and edition-aware organization.
My role
Product design & development
The scope
Android · Kotlin · Jetpack Compose
Project status
Native Android application
The project, at a glance
- The challenge
- A collector needs to distinguish editions, track ownership and recognize a duplicate, beyond simply saving a photo.
- My contribution
- I designed and built the native Android experience across camera capture, visual recognition, collection views and detailed records.
- What came out of it
- A working native app with documented product flows and store-listing material.
Native Android / CanDexSmall objects.
Small objects.
Entire worlds.
01Capture the object02Keep its story03Explore the collection
Original Android application · Demo collection
CanDexOriginal app / demo collection
01 / The question behind the work
A collection you can see, search and make your own.
The problem
A photo folder does not tell a collector which edition they own, what is missing, or whether a new find is a duplicate. Similar fronts can hide different sizes, countries and barcodes.
My contribution
Product flow, Android interface, visual collection system and native implementation across capture, cataloging, discovery and ownership controls.
What the work delivers
Kotlin / Jetpack Compose application, CameraX capture flow, ML Kit segmentation and barcode integration, Room-backed collection model, Collection and edition interfaces, Play Store presentation assets.
02 / Design decisionsThe thinking
The thinking
in the details.
Specific choices, connected to the work you can see.
Keep recognition editable
Combine barcode and visual evidence with optional back-label photos. Duplicate warnings retain a save-anyway path.
↳ Documented product behaviorStart locally
Keep the core collection on the device without requiring an account. Offer backup and a one-time Pro upgrade as separate decisions.
↳ Documented product behavior03 / Connecting the experience
How the pieces connect.
Select a stage to explore its purpose.
The moment
Photograph the can and collect barcode information.
The design intention
Let the object become the starting point.
New portfolio explanation based on the documented implementation and original project material.
04 / Inside the experienceThe details.
The details.
With a purpose.
Selected views, framed to explain the work. Complete original artifacts are available in the source collection below.
Browse through the objects
The grid foregrounds each can’s image and identity. This is the original Android application with a demo collection.
Change the view, keep the collection
The shelf offers a different way to browse the same objects while preserving their identity and ordering.
Give an object its full story
The detail screen connects imagery with edition metadata and ownership information.
Treat the edition as part of the object
The original metadata view preserves details that an image alone cannot communicate, including the product edition and recognition information.
Make room for more than one collection
The original collection switcher gives different groups of objects their own place. The capture uses demo collection content.
Original presentation boards6 source artifacts +
The complete originals behind this case study. Open a board to inspect the source presentation.
05 / What this work brings togetherCanDex
CanDex
in perspective.
The result
A working native app with store-listing material and documented product flows. Installation, revenue and conversion outcomes were not established in this review.
Sources & presentation notes +
Screens contain demo fixtures, not adoption evidence. This is computer-vision-assisted cataloging; do not imply a custom-trained model or perfect recognition. Current public store availability was not independently checked. This material is prepared for local portfolio review from the private project; source code is not included.
The journey diagrams and editorial explanations were prepared for this portfolio from the available project evidence. They are not additional historical research or measured outcomes.
Keep exploring / Next project
A studio for extracting a reusable style profile from reference images, editing prompt templates and producing structured batches of AI visuals.