Confidence-Aware Ensemble and Long-Word Refinement for Artistic Text Recognition

2026-08-30

AI Summary

本文针对艺术文字识别难题，提出一种基于置信度的集成方法和长词细化策略，提高了WordArt-V1.5基准上的识别准确率。

Abstract

Artistic Text Recognition (ATR) remains challenging because word images often combine decorative fonts, curved layouts, object-like characters, clutter, and severe distortions. This paper studies WordArt-V1.5 as a standardized benchmark for this setting and evaluates recent scene and artistic text recognizers under a common protocol. We propose a confidence-aware ensemble that combines SVTRv2, PARSeq, and MAERec after fine-tuning on the official training split. The ensemble selects predictions using the minimum confidence over disagreement positions, emphasizing characters that separate competing hypotheses. For long words, where a single character error can invalidate the whole prediction, we add a targeted refinement stage based on Needleman-Wunsch alignment and lexicon-guided correction. On the WordArt-V1.5 Test B split, the proposed system reaches 89.90% Word Recognition Accuracy, improving the best individual fine-tuned model by 1.77 percentage points. The long-word refinement produces a modest global gain, but improves the targeted long-word subset by 2.72 percentage points. Finally, an error analysis of all remaining mistakes shows that 48.8% are associated with labeling issues, visual ambiguity, or illegible samples, highlighting the value of diagnostic reporting for future ATR benchmarks and models. Our source code is available at https://github.com/lucas-azdias/Artistic-Text-Recognition/.

Problem

• Artistic Text Recognition
• decorative fonts
• curved layouts
• object-like characters
• clutter

Innovation

• confidence-aware ensemble
• long-word refinement
• Needleman-Wunsch alignment