Image Classification

A guide to AutoML image classification models

Overview

Image classification models can be used to categorize images and videos. Some popular use cases for this model type are object classification, brand identification, and vision moderation.

Which Base Model Should I Use?

Base ModelUse CaseHive ClassesDataset Requirements
Image Classification v2Well-suited for most image classification tasks--Minimum 10 examples per class
Visual Moderation v2Best for moderation tasks—leverages pre-trained Hive moderation labelsIncludedMinimum 50 examples per class

Labels and Classes for Hive Visual Moderation

If you select the Hive Visual Moderation base model, the model output will always contain all Hive Visual Moderation categories. A full list of moderation labels can be found in the Visual Moderation documentation.