ChartZero: Synthetic Priors Enable Zero Shot Chart Data Extraction

2026-05-07

AI Summary

This work addresses the challenges of poor generalization, fragmented curve extraction, and semantic mismatches between legends and data series in automatic line chart parsing—issues exacerbated by diverse visual styles and scarce real-world annotations. To overcome these limitations, the authors propose a zero-shot, end-to-end parsing framework trained exclusively on synthetic charts generated from mathematical functions, eliminating the need for real labeled data. The approach introduces a novel training paradigm driven by synthetic priors and incorporates a Global Orthogonal Instance (GOI) loss to mitigate curve fragmentation. Furthermore, it leverages open-vocabulary vision-language models to enable robust legend-to-series alignment. Evaluated on a newly established end-to-end reconstruction benchmark, the method substantially outperforms existing techniques, achieving highly generalizable unsupervised chart digitization.

Abstract

Automated data extraction from line charts remains fundamentally bottlenecked by extreme stylistic diversity and a severe scarcity of comprehensively annotated, real-world datasets. Current end-to-end pipelines depend heavily on costly manual annotations, crippling their ability to generalize across arbitrary aesthetics and grid layouts. Furthermore, existing models suffer from two critical failure modes during reconstruction. First, extracting thin, intersecting curves frequently causes structural fragmentation and the erasure of fine visual details, as standard architectures struggle against complex backgrounds. Second, semantic association is notoriously error-prone; current pipelines rely on rigid spatial heuristics that easily break down against the unpredictable legend placements of in-the-wild charts. Finally, measuring true progress is hindered by evaluation protocols that assess isolated sub-tasks rather than holistic, end-to-end data reconstruction. To address these foundational issues, we introduce ChartZero, a parsing framework that leverages synthetic priors to enable robust zero-shot chart data extraction. By training exclusively on a purely synthetic dataset of simple mathematical functions, our model completely bypasses the real-world annotation bottleneck. We overcome curve fragmentation via a novel Global Orthogonal Instance (GOI) loss, and replace brittle spatial rules with an open-vocabulary, Vision-Language Model (VLM)-guided legend matching strategy. Accompanied by a new metric and benchmark specifically designed for full end-to-end reconstruction, our evaluations demonstrate that ChartZero significantly advances generalized plot digitization without requiring real-world supervision. Code and dataset will be released upon acceptance.

Problem

• chart data extraction
• zero-shot learning
• stylistic diversity
• semantic association
• curve fragmentation

Innovation

• zero-shot chart extraction
• synthetic priors
• Global Orthogonal Instance loss
• Vision-Language Model
• end-to-end reconstruction