3D Virtual Try-On

Overview

3D Virtual Try-On enables shoppers to see how clothes, glasses or accessories look on them before buying. From a photo or live camera, the platform builds a 3D body or face model, fits products on it in real time and lets users turn, compare sizes and share the result. The solution is intended to help online stores increase conversion and reduce returns.

Objectives

• Build the body and face tracking and 3D reconstruction pipeline.
• Render products on the avatar in real time in the browser.
• Integrate the try-on widget into an online store.

Features

• Body and face pose estimation from a photo or live camera.
• 3D avatar generation using the user's measurements.
• Real-time garment and accessory fitting.
• Size recommendation and 360-degree preview.
• Web plugin for e-commerce stores.

Technologies and profile

Python, PyTorch, MediaPipe, Three.js, WebGL, React JS, FastAPI and Docker. Profile: AI / Computer vision engineering. Type: Computer Vision / 3D. One to two interns are planned.

Application documents

Attach a CV in PDF format. A short cover letter, GitHub or portfolio link and university internship agreement may also be provided.