Vinyl Javier

Wordnet for ImageNet Challenge (Vinyl Edition)

Description: Wordnet, created in 1985, is a hierarchical taxonomy that describes the world. Inspired in theories of human semantic memory developed in the late 1960s, in Wordnet, nouns, verbs, adjectives and adverbs are grouped into synonyms sets or synsets, each expressing a distinct concept. ImageNet is an image dataset organized according to the WordNet 3.0 nouns hierarchy, in which each node is depicted by thousands of images. From 2010 until 2017, the ImageNet Object Recognition Challenge (ILSVRC) was a key benchmark in object category classification, having a major impact in the increase of performance of Deep Convolutional Neural Networks.

Wordnet for ImageNet Challenge (Vinyl Edition) contains the 1000 synsets used in this challenge recorded in the highest sound quality that this analog format allows. This work highlights the relevance of the datasets used to train the AI models that run on the systems that we used on a daily basis, some of them inheriting classifications conceived more than 30 years ago, and invites us to thoughtfully analyse them.