RZCR: Zero-shot Character Recognition via Radical-based Reasoning

TL;DR

RZCR introduces a radical-based reasoning framework for zero-shot character recognition, effectively recognizing rare characters by decomposing characters into radicals and reasoning with a knowledge graph, addressing long-tail data distribution issues.

Contribution

The paper proposes a novel radical-based zero-shot recognition framework combining visual radical extraction and knowledge graph reasoning, improving recognition of few-sample characters.

Findings

RZCR outperforms existing methods on few-sample character datasets.

The radical decomposition approach enhances recognition of rare characters.

Knowledge graph reasoning improves accuracy in zero-shot scenarios.

Abstract

The long-tail effect is a common issue that limits the performance of deep learning models on real-world datasets. Character image datasets are also affected by such unbalanced data distribution due to differences in character usage frequency. Thus, current character recognition methods are limited when applied in the real world, especially for the categories in the tail that lack training samples, e.g., uncommon characters. In this paper, we propose a zero-shot character recognition framework via radical-based reasoning, called RZCR, to improve the recognition performance of few-sample character categories in the tail. Specifically, we exploit radicals, the graphical units of characters, by decomposing and reconstructing characters according to orthography. RZCR consists of a visual semantic fusion-based radical information extractor (RIE) and a knowledge graph character reasoner (KGR). RIE aims to recognize candidate radicals and their possible structural relations from character images in parallel. The results are then fed into KGR to recognize the target character by reasoning with a knowledge graph. We validate our method on multiple datasets, and RZCR shows promising experimental results, especially on few-sample character datasets.