DocILE Benchmark for Document Information Localization and Extraction

1 code implementation • 11 Feb 2023 • Štěpán Šimsa, Milan Šulc, Michal Uřičář, Yash Patel, Ahmed Hamdi, Matěj Kocián, Matyáš Skalický, Jiří Matas, Antoine Doucet, Mickaël Coustaty, Dimosthenis Karatzas

This paper introduces the DocILE benchmark with the largest dataset of business documents for the tasks of Key Information Localization and Extraction and Line Item Recognition.

Key Information Extraction Unsupervised Pre-training

Business Document Information Extraction: Towards Practical Benchmarks

no code implementations • 20 Jun 2022 • Matyáš Skalický, Štěpán Šimsa, Michal Uřičář, Milan Šulc

Information extraction from semi-structured documents is crucial for frictionless business-to-business (B2B) communication.

Artificial Dummies for Urban Dataset Augmentation

1 code implementation • 15 Dec 2020 • Antonín Vobecký, David Hurych, Michal Uřičář, Patrick Pérez, Josef Šivic

This is achieved with a data generator (called DummyNet) with disentangled control of the pose, the appearance, and the target background scene.

Autonomous Driving