TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/What Is Wrong With Scene Text Recognition Model Comparison...

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

Jeonghun Baek, Geewook Kim, Junyeop Lee, Sungrae Park, Dongyoon Han, Sangdoo Yun, Seong Joon Oh, Hwalsuk Lee

2019-04-03ICCV 2019 10Scene Text RecognitionImage Matching
PaperPDFCodeCodeCodeCode(official)CodeCodeCodeCodeCodeCodeCodeCodeCode

Abstract

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.

Results

TaskDatasetMetricValueModel
Scene ParsingSVTAccuracy87.5Baek et al.
Scene ParsingICDAR2015Accuracy71.8Baek et al.
Scene ParsingICDAR 2003Accuracy94.4Baek et al.
Scene ParsingICDAR2013Accuracy92.3Baek et al.
2D Semantic SegmentationSVTAccuracy87.5Baek et al.
2D Semantic SegmentationICDAR2015Accuracy71.8Baek et al.
2D Semantic SegmentationICDAR 2003Accuracy94.4Baek et al.
2D Semantic SegmentationICDAR2013Accuracy92.3Baek et al.
Scene Text RecognitionSVTAccuracy87.5Baek et al.
Scene Text RecognitionICDAR2015Accuracy71.8Baek et al.
Scene Text RecognitionICDAR 2003Accuracy94.4Baek et al.
Scene Text RecognitionICDAR2013Accuracy92.3Baek et al.

Related Papers

Linguistics-aware Masked Image Modeling for Self-supervised Scene Text Recognition2025-03-24Efficient and Accurate Scene Text Recognition with Cascaded-Transformers2025-03-24Accurate Scene Text Recognition with Efficient Model Scaling and Cloze Self-Distillation2025-03-20A Context-Driven Training-Free Network for Lightweight Scene Text Segmentation and Recognition2025-03-19EventSTR: A Benchmark Dataset and Baselines for Event Stream based Scene Text Recognition2025-02-13Billet Number Recognition Based on Test-Time Adaptation2025-02-13Ocean-OCR: Towards General OCR Application via a Vision-Language Model2025-01-26Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance2024-12-13