Flaws of ImageNet, Computer Vision's Favourite Dataset
ImageNet-1k suffers from incorrect labels, overlapping classes, domain shifts, and duplicates that undermine its reliability as a computer vision benchmark.
Published Nov 26, 20241 citationarXiv ↗
Only vote on papers you've read. Sign in to vote.
Abstract
Since its release, ImageNet-1k dataset has become a gold standard for evaluating model performance. It has served as the foundation for numerous other datasets and training tasks in computer vision. As models have improved in accuracy, issues related to label correctness have become increasingly apparent. In this blog post, we analyze the issues in the ImageNet-1k dataset, including incorrect labels, overlapping or ambiguous class definitions, training-evaluation domain shifts, and image duplicates. The solutions for some problems are straightforward. For others, we hope to start a broader conversation about refining this influential dataset to better serve future research.