Embroidery Automation and Optimization for Mass Customization

Embroidery is a very popular art and a method of customization in the fashion industry. The method of embroidery has changed little since the advent of programmable sewing machines in the 20th century. Today, there is high demand for mass customization: consumers want their personalized embroidery patterns on their clothing. Technology already allows consumers to submit a picture while placing an order for clothing, and to receive the clothing with an embroidered pattern matching the picture. However, this process is time-consuming and human labour-intensive: a human worker takes the picture, picks out the thread colors, and creates a sequence of embroidery-machine instructions to create the pattern. We propose to use machine learning to automate and optimize this process. We will develop algorithms to pick the appropriate thread colors, generate previews of the final embroidered pattern, and minimize the stitch count subject to symmetry and quality constraints.

Faculty Supervisor:

Jia Yuan Yu

Student:

Partner:

Monsieur Stitches Inc

Discipline:

Computer science

Sector:

Retail trade

University:

Concordia University

Program: