๐Ÿ“œ
Developer Documentation
  • ๐Ÿ‘‹EyePop.ai Introduction
  • ๐ŸŽฏGetting Started
    • ๐Ÿ‘จโ€๐Ÿ’ปPop Quick Start
    • ๐Ÿ’ชLow Code Examples
  • ๐Ÿ—๏ธAPI Key
  • ๐Ÿท๏ธFinding People & Objects
  • SDKs
    • โ˜๏ธReact/Node SDK
      • Render 2D (Visualization)
      • Composable Pops
    • ๐ŸPython SDK
      • Composable Pops
  • Self Service Training
    • Dataset SDK (Node)
    • ๐Ÿ‹๏ธHow To Train a Model
      • Defining Your Computer Vision Model
      • Example Use Case: Detecting Eyeglasses
      • Preparing & Uploading Data
      • Using EyePop.aiโ€™s AutoLabeler
      • Human Review
      • Data Augmentation Setup
      • Training in Progress
      • Deployment
        • Deploy to Windows Runtime
      • Previewing Results
      • Iterative Training
      • Deep Dives (FAQ)
  • EyePop.ai Visual Intelligence
    • Reports
Powered by GitBook
On this page
  1. Self Service Training
  2. How To Train a Model

Example Use Case: Detecting Eyeglasses

This example walks you through creating and training a model that identifies whether eyeglasses appear in an advertising image.

Defining your Problem

In this case weโ€™re simply looking for clear images of glasses in various contexts. We also want to know how much space in the frame they take up, if a person is wearing them, and how centered they are in the frame.

To begin, navigate to the My Models page, click on Create a New Model, and choose Find Objects as your model type. Name your model and define the label for detection, eg., โ€œeyeglassesโ€. When you click Continue, youโ€™ll be able to upload an image dataset on the right hand side of the screen. The next step walks you through preparing data to make sure you have a representative dataset for training your model.

PreviousDefining Your Computer Vision ModelNextPreparing & Uploading Data

Last updated 5 months ago

๐Ÿ‹๏ธ