Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Inception score
open-in-new
<p>The Inception Score (IS) is an algorithm used to assess the quality of images created by a <a href="/facts/Generative_model/JUgExNIP">generative</a> image model such as a <a href="/facts/Generative_adversarial_network/SGNuGrlo">generative adversarial network</a> (GAN). The score is calculated based on the output of a separate, pretrained <a href="/facts/Inception_(deep_learning_architecture)/lBx4Uv5q">Inception v3</a> image classification model applied to a sample of (typically around 30,000) images generated by the generative model. The Inception Score is maximized when the following conditions are true: </p> <ol><li>The <a href="/facts/Entropy_(information_theory)/NLg4NLvt">entropy</a> of the distribution of labels predicted by the Inceptionv3 model for the generated images is minimized. In other words, the classification model confidently predicts a single label for each image. Intuitively, this corresponds to the desideratum of generated images being "sharp" or "distinct".</li> <li>The predictions of the classification model are evenly distributed across all possible labels. This corresponds to the desideratum that the output of the generative model is "diverse".</li></ol> <p>It has been somewhat superseded by the related <a href="/facts/Fr%25C3%25A9chet_inception_distance/VvAfWdGX">Fréchet inception distance</a>. While the Inception Score only evaluates the distribution of generated images, the FID compares the distribution of generated images with the distribution of a set of real images ("ground truth"). </p>