Toward Zero-shot Character Recognition: A Gold Standard Dataset with Radical-level Annotations

TL;DR

This paper introduces ACCID, a comprehensive ancient Chinese character dataset with radical and character annotations, and proposes a baseline zero-shot OCR method leveraging radical decomposition, addressing the lack of benchmarks and annotations.

Contribution

The paper creates ACCID, a novel dataset with radical-level annotations, and develops a baseline zero-shot OCR approach using radical decomposition and data augmentation techniques.

Findings

ACCID dataset effectively supports zero-shot OCR research.

The baseline method achieves promising results on radical recognition.

Synthetic data augmentation improves OCR model robustness.

Abstract

Optical character recognition (OCR) methods have been applied to diverse tasks, e.g., street view text recognition and document analysis. Recently, zero-shot OCR has piqued the interest of the research community because it considers a practical OCR scenario with unbalanced data distribution. However, there is a lack of benchmarks for evaluating such zero-shot methods that apply a divide-and-conquer recognition strategy by decomposing characters into radicals. Meanwhile, radical recognition, as another important OCR task, also lacks radical-level annotation for model training. In this paper, we construct an ancient Chinese character image dataset that contains both radical-level and character-level annotations to satisfy the requirements of the above-mentioned methods, namely, ACCID, where radical-level annotations include radical categories, radical locations, and structural relations. To increase the adaptability of ACCID, we propose a splicing-based synthetic character algorithm to augment the training samples and apply an image denoising method to improve the image quality. By introducing character decomposition and recombination, we propose a baseline method for zero-shot OCR. The experimental results demonstrate the validity of ACCID and the baseline model quantitatively and qualitatively.