A Hybrid Vision Transformer Approach for Mathematical Expression Recognition

TL;DR

This paper focuses on the part of math formula that has 2-D spatial structure relationship, and how to identify the spatial relationships between all characters of the math formula.

Abstract

One of the crucial challenges taken in document analysis is mathematical expression recognition. Unlike text recognition which only focuses on one-dimensional structure images, mathematical expression recognition is a much more complicated problem because of its two-dimensional structure and different symbol size. In this paper, we propose using a Hybrid Vision Transformer (HVT) with 2D positional encoding as the encoder to extract the complex relationship between symbols from the image. A coverage attention decoder is used to better track attention's history to handle the under-parsing and over-parsing problems. We also showed the benefit of using the [CLS] token of ViT as the initial embedding of the decoder. Experiments performed on the IM2LATEX-100K dataset have shown the effectiveness of our method by achieving a BLEU score of 89.94 and outperforming current state-of-the-art methods.

Figures (6)

• Figure 1: A general overview of our proposed framework
• Figure 2: The proposed architecture pipeline and illustration of ViT block.
• Figure 3: Comparison between our proposed method and the baseline model at different sequence length.
• Figure 4: Examples of self-attention map of [CLS] token embedding.
• Figure 5: Visualization of a step-by-step decoding process using our method on example math expression image.