An evolving system for understanding your wardrobe and personal style.
Muse explores how clothing can become something a person can understand, organize, and interact with digitally — turning a physical wardrobe into a structured representation that can support real decisions about what to wear.
Exploring and developing the foundations of an AI-powered wardrobe and personal-style system.
The software and technical foundations needed to capture, understand, and work with clothing digitally.
How can clothing photographed by a person be transformed into useful, structured information that supports meaningful wardrobe and outfit decisions?
Continue turning the Muse concept into a working, testable product experience.
A wardrobe contains more information than the clothes in it.
A person may own dozens or hundreds of pieces, but the relationships between them — color, style, type, fit, context, compatibility, frequency of use, personal preference — are difficult to see.
Muse explores whether AI can make those relationships visible. The long-term direction is a personal system that understands the clothes someone owns and uses that understanding to help them navigate their wardrobe and their style.
From "what clothes do I own?" toward "what works together, and what should I wear?"
Understand
Digitize clothing and extract useful information from images.
Organize
Build a structured representation of a person's wardrobe.
Evaluate
Explore ways of describing and assessing individual pieces.
Combine
Identify relationships between garments and possible combinations.
Create
Generate complete outfit possibilities from the available wardrobe.
Recommend
Help answer the practical question: what should I wear?
Learn
Gradually build an understanding of an individual's style and preferences.
-
Sept 2026
Computer vision and product recognition exploration
Google Cloud's vision capabilities and Product Search concepts were investigated for clothing imagery, including a product-set approach and a Google Cloud environment set up for the work.
-
Sept 2026
AI wardrobe intelligence direction
The product direction was articulated more broadly: understanding clothing, evaluating pieces, identifying combinations, creating outfits, recommending what to wear, identifying gaps, and learning personal style.
-
Aug 2026
Muse established as a fashion application concept
Muse emerged as a fashion and clothing application exploring how technology could understand a person's wardrobe and style.
Questions currently being investigated.
Clothing recognition
ExploredHow accurately can a garment be recognized and described from a photograph taken by a person?
Visual product search
ExploredCan product-search infrastructure connect a photographed garment to useful product information?
Wardrobe → outfit matching
ProposedOnce garments are represented together, what makes two pieces genuinely compatible?
Experiments are recorded in full in the Lab. Nothing here is presented as an implemented capability.
Muse is being developed as an evolving system, not a fixed specification.
Product
The concept, experience, features and product thinking.
Development
The software, architecture, image processing, recognition and AI work.
Software
Working software, prototypes and technical tools built for Muse.
Experiments
Small investigations used to answer technical and product questions.
Decisions
A record of important choices and the reasoning behind them.
Open questions
The things Muse does not yet know how to answer.
Project log
A chronological record of how the project evolves.
Roadmap
Where the project is going next.
Features & status
What is built, explored, planned and still only a direction.
Status is stated, not implied.
Muse is an active exploration, so the site distinguishes carefully between what exists and what is intended. Every capability, experiment and roadmap item carries a status.
Actually implemented.
Currently being implemented.
Experimentally evaluated.
Investigated, not necessarily implemented.
Intentionally identified future work.
An idea not yet committed to.
Longer-term possibility.
Not resolved.