Leveraging Computer Vision Priciples for Image-Based Shoe Recommendation

Abstract

We have developed an image-based shoe recommendation system to help modernize the retail industry. This system assesses a dataset of over 10,000 men’s shoes and extracts the shape, color, and texture associated with each of these shoes. Using a combination of contour detection, scale-invariant feature transform, color-based clustering, and several other… (More)

14 Figures and Tables

Topics

  • Presentations referencing similar topics