9 citations · 23 across the 16 of their papers we have counts for
4 papers · 1 filter
Label Errors in the Tobacco3482 Dataset
Gordon Lim, Stefan Larson, Kevin Leach
Tobacco3482 is a widely used document classification benchmark dataset. However, our manual inspection of the entire dataset uncovers widespread ontological issues, especially larg…
ShabbyPages: A Reproducible Document Denoising and Binarization Dataset
Alexander Groleau, Kok Wei Chee, Stefan Larson +2
Document denoising and binarization are fundamental problems in the document processing space, but current datasets are often too small and lack sufficient complexity to effectivel…
Evaluating Out-of-Distribution Performance on Document Image Classifiers
Stefan Larson, Gordon Lim, Yutong Ai +2
The ability of a document classifier to handle inputs that are drawn from a distribution different from the training distribution is crucial for robust deployment and generalizabil…
Augraphy: A Data Augmentation Library for Document Images
Alexander Groleau, Kok Wei Chee, Stefan Larson +2
This paper introduces Augraphy, a Python library for constructing data augmentation pipelines which produce distortions commonly seen in real-world document image datasets. Augraph…