cs.CVAug 6, 2026

Bar-JEPA: Extracting Values from Bar Chart with Joint-Embedding Predictive Architecture

Authors: Poonam PoonamAlexander EppleTimo Ropinski

Organizations: Institute of Media Informatics, Visual Computing Group, Ulm University, Germany

Abstract

Bar charts are commonly used in data visualization, and while they are easily understood by humans, it is non-trivial to extract the underlying data computationally. For a machine-learning-based approach, training chart de-rendering models usually requires labeled, real-world data. Labeling data is a time consuming task, which is why annotated data is scarce. Models can learn more efficiently when provided with features of high semantic quality, which a joint-embedding predictive architecture (JEPA) is designed to learn in a self-supervised manner. We present a per-bar, numerical value recovery pipeline for bar charts, where a JEPA encoder is used to produce semantically rich latent features. The decoder model consuming these features is simple and quick to train and outputs the coordinates of ticks and bars, which can be used to recover bar values. The effectiveness of self-supervised finetuning and quality of the extracted features is evident when comparing our model to end-to-end supervised baselines. Code, datasets and checkpoints are available on \href{https://github.com/dralois/Bar-JEPA}{GitHub}.

Explore similar work

CardsList
  1. Self-Ensembling Vision-Language Models for Chart Data Extraction

    May 26, 2026Thomas Berkane, Qianyi Wang, Maimuna S. MajumderChartTabular Data

  2. ChartZero: Synthetic Priors Enable Zero Shot Chart Data Extraction

    May 7, 2026Md Touhidul Islam, Yasir Mahmud, Sujan Kumar Saha +2ChartPrior Knowledge