{"url":"/dataset/icon645","name":"Icon645","full_name":null,"description_markdown":"Icon645 is a large-scale dataset of icon images that cover a wide range of objects:\r\n\r\n* **645,687** colored icons\r\n* **377** different icon classes\r\n\r\nThese collected icon classes are frequently mentioned in the [IconQA](https://iconqa.github.io/) questions. In this work, we use the icon data to pre-train backbone networks on the icon classification task in order to extract semantic representations from abstract diagrams in IconQA. On top of pre-training encoders, the large-scale icon data could also contribute to open research on abstract aesthetics and symbolic visual understanding.","description_withheld":null,"homepage":"https://iconqa.github.io/","introduced_date":"2021-10-25","introduced_date_note":null,"introduced_by":{"paper":"/paper/iconqa-a-new-benchmark-for-abstract-diagram","title":"IconQA: A New Benchmark for Abstract Diagram Understanding and Visual Language Reasoning","first_author":"Pan Lu","url":null},"license":{"name":"CC BY-NC-SA 4.0","url":"https://github.com/lupantech/InterGPS/blob/main/LICENSE"},"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Image Classification","url":"/task/image-classification","datasets_with_task":"/datasets/task/image-classification"},{"name":"Unsupervised Pre-training","url":"/task/unsupervised-pre-training","datasets_with_task":"/datasets/task/unsupervised-pre-training"}],"languages":[],"variants":["Icon645"],"data_loaders":[],"num_papers_in_archive":1,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[],"papers_with_a_benchmark_row":[],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}