Body Measurement App Using Machine Learning Technology

With interests spanning prepaid vouchers, clothing, and retail, the client focuses on developing B2C applications for various platforms. Recently, the client engaged with a customer in the school uniform tailoring business who needed an automated solution to speed up the labor-intensive task of measuring each individual child. This led to the concept of a body measurement app designed to streamline the process.

The Challenge

Tailoring often involves manually measuring each individual, a time-consuming and intrusive process, particularly at scale. The client required a less intrusive, automated solution to estimate measurements while maintaining privacy, enabling faster production of tailored uniforms.

Recognizing the broader application of such a solution, the client sought to develop a body measurement app that could automate uniform-size measurements for school children, with potential uses across various industries. However, lacking the technical expertise to create a proof of concept (POC), they approached Experion for help in leveraging artificial intelligence and machine learning to develop this innovative solution.

Our Solution

Experion proposed a three-step solution for the body measurement app, involving:

Scanning of the body: A full-body image scan using a true depth camera to render a 1:1 scale watertight 3D mesh estimate of the body.
Identification of the key points: Using an intelligent algorithm, key points on the body (shoulders, waist, chest, etc.) were identified based on ISO-8559-1 definitions for standardized body measurements in the clothing industry.
Estimation of Body Measurements: : Once the key points were identified, mathematical equations were applied to the mesh coordinates, providing accurate measurements for tailoring.

The algorithm, developed in Python 3.6 using computer vision, deep learning, and scientific computation libraries, processed the depth sensor-generated mesh input. The app displayed the estimated key points and measurements on a graphical interface for customer verification or further adjustments.

This innovative solution offered contactless, automated body size measurements, enhancing accuracy and reducing time and effort, while opening up opportunities for various applications such as foot-size determination and medical diagnostics.

Why Us

Tech Stack

Language: Python 3.6
Mesh processing: Trimesh, Meshlab, Scipy, Numpy
Mesh Generation: Makehuman
Machine Learning Algorithms: KDtree, K Means Clustering, Random Forest
UI: React JS, Typescript

Business Impact

Successful POC for a breakthrough body measurement app that could revolutionize the retail tailoring business.
Contactless body-size measurement ensuring social distancing and full privacy.
Batch processing and automation reduce measurement time, increasing business volume with minimal supervision.
Secure digital storage of measurements enables easy reuse for future orders.

Contact

Case Studies