Summary of Local and Global Graph Modeling with Edge-weighted Graph Attention Network For Handwritten Mathematical Expression Recognition, by Yejing Xie et al.
Local and Global Graph Modeling with Edge-weighted Graph Attention Network for Handwritten Mathematical Expression Recognition
by Yejing Xie, Richard Zanibbi, Harold Mouchère
First submitted to arxiv on: 24 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: 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 The paper introduces a novel approach to Handwritten Mathematical Expression Recognition (HMER) using graph-based modeling techniques. The proposed End-to-end model with an Edge-weighted Graph Attention Mechanism (EGAT) simultaneously classifies nodes and edges, allowing for the prediction of symbol classes and their relationships within mathematical expressions. Additionally, the paper proposes stroke-level Graph Modeling methods for local (LGM) and global (GGM) information, applying an end-to-end model to online HMER tasks. The system demonstrates superior performance in symbol detection, relation classification, and expression-level recognition. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Handwritten Mathematical Expression Recognition is a way for computers to understand handwritten math problems. This paper shows how to improve this process by using special graph-based models. These models are good at recognizing patterns and relationships between different parts of the math problem. The new approach uses two main techniques: one for local details and another for global structure. This helps the computer understand the whole expression better. The results show that this method is very effective in detecting symbols, understanding relationships, and recognizing entire expressions. |
Keywords
* Artificial intelligence * Attention * Classification
