Clothes detection and AI outfit recommendations
We built a step-by-step “personal stylist”: computer vision that detects garments and accessories in user photos, analyzes style and recommends outfits.
- Industry
- Retail, Fashion
- Services
- Computer vision, Recommendation systems, Full-stack development
- Duration
- 5 months
- Project duration
- 5 months
- Specialists on the team
- 5
Intro
AI has automated parts of many professions. Our brief was to build a step-by-step personal stylist that works from a user’s photos and preferences.
The system detects and recognizes garments and accessories, analyzes style — colors, shapes and fabric — and recommends outfits based on user preferences and inferred style features.
Goals and challenges
- 01
Accurate detection
Garments had to be detected accurately on real user photos.
- 02
High load
The system had to handle high load at scale.
- 03
Easy deployment
It had to be straightforward to deploy.
Solution
We delivered an end-to-end computer vision and recommendation stack with scalable deployment on AWS.
Team
- 2 ML Engineers
- 1 Front-end Developer
- 1 UX/UI Designer
- 1 DevOps Engineer
Project details
- Industry
- Retail, Fashion
- Services
- Computer vision, Recommendation systems, Full-stack development
- Technologies
- Python
- TensorFlow
- React
- AWS
- aiohttp
- REST
- Visual search
- Team
- 5 — 2 ML engineers, 1 front-end developer, 1 UX/UI designer, 1 DevOps engineer
- Duration
- 5 months
Discover more
How we sped up public transit routing around the globe
Public Transit · Routing · OpenTripPlanner · Performance
AI-powered construction monitoring platform
Construction · Computer Vision · VLM · Drones · 3D Visualization · SaaS
Faster APIs, automation and AI recommendations for WKO Inhouse
Gov · App development · Recommendation System · OpenAI · Data Science