Modular Multimodal Architecture for Document Classification
arXiv:1912.04376
Abstract
Page classification is a crucial component to any document analysis system, allowing for complex branching control flows for different components of a given document. Utilizing both the visual and textual content of a page, the proposed method exceeds the current state-of-the-art performance on the RVL-CDIP benchmark at 93.03% test accuracy.
Cited by in corpus (5)
- LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding
- Document AI: Benchmarks, Models and Applications
- StructuralLM: Structural Pre-training for Form Understanding
- Towards a Multi-modal, Multi-task Learning based Pre-training Framework for Document Representation Learning
- SelfDoc: Self-Supervised Document Representation Learning