Skip to yearly menu bar Skip to main content


Poster
in
Datasets and Benchmarks: Dataset and Benchmark Poster Session 3

OmniPrint: A Configurable Printed Character Synthesizer

Haozhe Sun · Wei-Wei Tu · Isabelle Guyon


Abstract:

We introduce OmniPrint, a synthetic data generator of isolated printed characters, geared toward machine learning research. It draws inspiration from famous datasets such as MNIST, SVHN and Omniglot, but offers the capability of generating a wide variety of printed characters from various languages, fonts and styles, with customized distortions. We include 935 fonts from 27 scripts and many types of distortions. As a proof of concept, we show various use cases, including an example of meta-learning dataset designed for the upcoming MetaDL NeurIPS 2021 competition. OmniPrint is available at https://github.com/SunHaozhe/OmniPrint.

Chat is not available.