Summary of Layereddoc: Domain Adaptive Document Restoration with a Layer Separation Approach, by Maria Pilligua et al.
LayeredDoc: Domain Adaptive Document Restoration with a Layer Separation Approach
by Maria Pilligua, Nil Biescas, Javier Vazquez-Corral, Josep Lladós, Ernest Valveny, Sanket Biswas
First submitted to arxiv on: 12 Jun 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel intelligent document processing system called LayeredDoc is introduced to overcome limitations of traditional methods in domain adaptability. The approach uses a text-graphic layer separation method that enhances the performance of document image restoration (DIR) systems by dynamically adjusting to input documents’ characteristics. This hierarchical DIR framework consists of two layers: one for coarse-grained graphic components and another for refining machine-printed textual content. Evaluation using the LayeredDocDB dataset, developed for this study, demonstrates strong generalization capabilities for the DIR task. The model is initially trained on a synthetically generated dataset and offers a promising solution for handling variability in real-world data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to make computers better at reading documents is introduced. This method, called LayeredDoc, helps machines understand different types of documents without needing lots of extra training. It does this by looking at both the pictures and text in a document, and then using that information to fix any mistakes. The researchers tested their approach with a new dataset they created, and it worked really well. |
Keywords
» Artificial intelligence » Generalization
