Style AI
𝙎𝙩𝙮𝙡𝙚 𝘼𝙄 is an innovative system that generates dresses through an interactive genetic algorithm. This service creates a dress for a specific occasion and style selected by the user. By answering questions and selecting their favourite dresses in each population, the user is guiding the algorithm into creating the perfect dress.
Timeline
10.2022-01.2023Project Course (MSc)
Computational Creativity @ Leiden University
Role (3 Members Team)
Research: Background & market analysisFlowchart & WireframingBrandingPrototyping (Hi-fi)Context
Generative AI is increasingly used in creative industries, yet the relationship between human creativity and algorithmic generation remains unclear.
In this MSc project, our team explored how AI could act not as a replacement for creativity, but as a collaborative design partner in custom fashion design.
Problem Definition
Fashion design is highly subjective, making it difficult to evaluate or automate through conventional AI systems.
Existing AI fashion tools often:
prioritize automation over collaboration
limit user creative control
produce outputs disconnected from user intent
core problem
How might we design an AI system that supports human creativity rather than overshadowing it?
Hypothesis
A genetic algorithm guided by iterative user preference selection would enable
meaningful human-AI co-creation while maintaining user agency.
Approach
During the research, we analysed 2 main AI approaches in fashion.
-
In this idea, AI suggests combinations from users’ existing wardrobe items. #mixandmatch
Rejected because:
heavy dependency on user-provided images
risk of plagiarism
limited generative novelty
-
AI designs entirely new clothing items beyond the user's current wardrobe. #new_item_generation
Chosen because:
higher originality potential
reduced dependence on direct inputs
stronger opportunity for exploration
→ This decision shaped both the algorithmic logic and UX design strategy.
solution
Foundational concept: Feature recombination model
This direction allows for higher novelty, reduces plagiarism concerns, and invites meaningful human-AI co-creation by integrating user preferences into an evolving design process.
The following shows the flowchart of the experience and the prototype.
The experience was designed around 3 principles:
Progressive preference input
Real-time algorithm adaptation
Balance between guidance and user control
User flow:
Select occasion
Refine preferences through iterative selection
Edit final details
Results
check the prototype
Evaluation
We developed our assessment metrics based on other researchers. The metrics are:
Independence
Variety and experimentation
Originality
Exploration
Expressiveness
Enjoyment
Collaboration
Satisfaction with final outcome
-
Cherry, E., & Latulipe, C. (2014). Quantifying the Creativity Support of Digital Tools through the Creativity Support Index. ACM Transactions on Computer-Human Interaction, 21(4), 1–25. doi: 10.1145/2617588
Jordanous, A. A. (2012). Standardised Procedure for Evaluating Creative Systems: Computational Creativity Evaluation Based on What it is to be Creative. Cognitive Computation 4, 246–279. doi: 10.1007/s12559-012-9156-1
Kantosalo, A., & Riihiaho, S. (2018). Experience evaluations for human–computer co-creative processes – planning and conducting an evaluation in practice. Connection Science, 31(1), 60–81. doi: 10.1080/09540091.2018.1432566
4.2 / 5
Overall satisfaction
Results suggest:
users reported enjoyable collaboration
high intention to reuse
Limitations: small sample size (n=5), exploratory interpretation only
Reflection
The project revealed a critical tension:
The algorithm sometimes influenced outcomes more strongly than the user intended.
Key takeaways:
co-creation requires careful balance between automation and autonomy
richer datasets are essential for creative diversity
evaluation metrics for creativity must be clearly defined early
Future improvements:
increase participant diversity
refine interaction model to strengthen user agency
expand dataset variability