# What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis

arXiv: https://arxiv.org/abs/1904.01906 · HTML: https://mlsanity.com/paper/1904.01906/

## Key facts

- Cited by 191 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Apr 2019.

## Built on by (most cited first, 30 of 191)

- [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://mlsanity.com/paper/2203.10244/) (2022)
- [OCRBench: On the Hidden Mystery of OCR in Large Multimodal Models](https://mlsanity.com/paper/2305.07895/) (2023)
- [OCR-free Document Understanding Transformer](https://mlsanity.com/paper/2111.15664/) (2021)
- [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://mlsanity.com/paper/2109.10282/) (2021)
- [Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition](https://mlsanity.com/paper/2103.06495/) (2021)
- [TextOCR: Towards large-scale end-to-end reasoning for arbitrary-shaped scene text](https://mlsanity.com/paper/2105.05486/) (2021)
- [Chart-to-Text: A Large-Scale Benchmark for Chart Summarization](https://mlsanity.com/paper/2203.06486/) (2022)
- [Scene Text Recognition with Permuted Autoregressive Sequence Models](https://mlsanity.com/paper/2207.06966/) (2022)
- [SVTR: Scene Text Recognition with a Single Visual Model](https://mlsanity.com/paper/2205.00159/) (2022)
- [Vision Transformer for Fast and Efficient Scene Text Recognition](https://mlsanity.com/paper/2105.08582/) (2021)
- [LaTr: Layout-Aware Transformer for Scene-Text VQA](https://mlsanity.com/paper/2112.12494/) (2021)
- [What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels](https://mlsanity.com/paper/2103.04400/) (2021)
- [Conditional Text Image Generation with Diffusion Models](https://mlsanity.com/paper/2306.10804/) (2023)
- [Benchmarking Chinese Text Recognition: Datasets, Baselines, and an Empirical Study](https://mlsanity.com/paper/2112.15093/) (2021)
- [Multi-modal In-Context Learning Makes an Ego-evolving Scene Text Recognizer](https://mlsanity.com/paper/2311.13120/) (2023)
- [PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering Network](https://mlsanity.com/paper/2104.05458/) (2021)
- [TextStyleBrush: Transfer of Text Aesthetics from a Single Example](https://mlsanity.com/paper/2106.08385/) (2021)
- [Rethinking Text Line Recognition Models](https://mlsanity.com/paper/2104.07787/) (2021)
- [Toward Understanding WordArt: Corner-Guided Transformer for Scene Text Recognition](https://mlsanity.com/paper/2208.00438/) (2022)
- [Revisiting Scene Text Recognition: A Data Perspective](https://mlsanity.com/paper/2307.08723/) (2023)
- [TextCtrl: Diffusion-based Scene Text Editing with Prior Guidance Control](https://mlsanity.com/paper/2410.10133/) (2024)
- [SynthTIGER: Synthetic Text Image GEneratoR Towards Better Text Recognition Models](https://mlsanity.com/paper/2107.09313/) (2021)
- [PIMNet: A Parallel, Iterative and Mimicking Network for Scene Text Recognition](https://mlsanity.com/paper/2109.04145/) (2021)
- [ABINet++: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Spotting](https://mlsanity.com/paper/2211.10578/) (2022)
- [OneChart: Purify the Chart Structural Extraction via One Auxiliary Token](https://mlsanity.com/paper/2404.09987/) (2024)
- [Exploring Stroke-Level Modifications for Scene Text Editing](https://mlsanity.com/paper/2212.01982/) (2022)
- [CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition](https://mlsanity.com/paper/2111.11011/) (2021)
- [Multi-Granularity Prediction for Scene Text Recognition](https://mlsanity.com/paper/2209.03592/) (2022)
- [Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features](https://mlsanity.com/paper/2111.15263/) (2021)
- [GLASS: Global to Local Attention for Scene-Text Spotting](https://mlsanity.com/paper/2208.03364/) (2022)

## Abstract (arXiv)

Many new proposals for scene text recognition (STR) models have been introduced in recent years. While each claim to have pushed the boundary of the technology, a holistic and fair comparison has been largely missing in the field due to the inconsistent choices of training and evaluation datasets. This paper addresses this difficulty with three major contributions. First, we examine the inconsistencies of training and evaluation datasets, and the performance gap results from inconsistencies. Second, we introduce a unified four-stage STR framework that most existing STR models fit into. Using this framework allows for the extensive evaluation of previously proposed STR modules and the discovery of previously unexplored module combinations. Third, we analyze the module-wise contributions to performance in terms of accuracy, speed, and memory demand, under one consistent set of training and evaluation datasets. Such analyses clean up the hindrance on the current comparisons to understand the performance gain of the existing modules.