LineEX: Data Extraction from Scientific Line Charts
V P Shivasankaran, Muhammad Yusuf Hassan, Mayank Singh
Abstract
In this paper, we introduce LineEX that extracts data from scientific line charts. We adapt existing vision transformers and pose detection methods and showcase significant performance gains over existing SOTA baselines. We also propose a new loss function and present its effectiveness against existing loss functions. In addition, we synthetically created the largest line chart dataset comprising 430K images. The code is available at https: //github.com/Shiva-sankaran/LineEX.
Topics & Concepts
Computer scienceChartLine (geometry)Code (set theory)TransformerSource lines of codeFunction (biology)Artificial intelligenceData miningSoftwareEngineeringProgramming languageMathematicsElectrical engineeringSet (abstract data type)VoltageBiologyEvolutionary biologyStatisticsGeometryImage and Object Detection TechniquesRobotics and Sensor-Based LocalizationAdvanced Image and Video Retrieval Techniques