1. Introduction
    1. A visual dataset of styles and substitutes
    2. Related work
    3. A visual and relational recommender system
  2. The Model
    1. Style space
    2. Personalizing styles to individual users
    3. Features
  3. Training
  4. Experiments
    1. Experimental protocol
    2. Personalized recommendations
  5. Visualizing Style Space
  6. Generating Recommendations
  7. Outfits In The Wild
  8. Conclusion

(SIGIR 2015) Image-based recommendations on styles and substitutes
Paper: https://cseweb.ucsd.edu/~jmcauley/pdfs/sigir15.pdf
Code: https://cseweb.ucsd.edu/~jmcauley/code/imageGraph.tar.gz

model this human sense of the relationships between objects based on their appearance.

capturing the largest dataset possible and developing a scalable method for uncovering human notions of the visual relationships within.

Introduction

A visual dataset of styles and substitutes

A visual and relational recommender system

The Model

Style space

Personalizing styles to individual users

Features

Training

Experiments

Experimental protocol

Personalized recommendations

Visualizing Style Space

Generating Recommendations

Outfits In The Wild

Conclusion