Mini-project
Retail, E-commerce & Consumer Goods
Fashion image tagging and similar-product search
Tag clothing images by type and colour and find similar products in a catalogue.
The brief
Shoppers want "more like this", and catalogue teams tag thousands of images by hand. Train an image model that tags clothing by type and colour, and use image embeddings to find visually similar products. Good looks like: tagging accuracy per class, a similarity search judged on test queries, and a demo.
What you will build
- Explore the product images and labels
- Fine-tune an image model for type and colour
- Build image embeddings and a similarity index
- Evaluate tagging and search quality
- Build a "more like this" demo
Free, open data to build it with
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.