Skip to content

Dataset · flwrlabs

Office and home object images in 4 styles

flwrlabs/office-home

A classic dataset with images of 65 object categories to evaluate computer vision adaptation to different domains.

What's inside

It contains images and text labels for 65 everyday object categories commonly found at home and in the office. The data is divided into 4 visual domains: artistic illustrations, cliparts, product photos, and real-world images. The total volume ranges from 10,000 to 100,000 examples.

What you can do with it

  • Evaluating and debugging domain adaptation algorithms in image classification
  • Training visual models to recognize objects regardless of image style
  • Benchmarking neural network robustness to data distribution shift
Size
10K–100K rows
Downloads this month
22.5K
Likes
2

How to load it

from datasets import load_dataset

ds = load_dataset("flwrlabs/office-home")