APEX-Net: Automatic Plot Extraction Network

Aalok Gangopadhyay, Prajwal Singh, Shanmuganathan Raman

Abstract: Automatic plot extraction involves understanding and inferring the data distribution and therefore, extracting individual line plots from an image containing multiple 2D line plots. It is an important problem having many real-world applications. The existing methods for addressing this problem involve a significant amount of human intervention. To minimize this intervention, we propose APEX-Net, a deep learning based framework with novel loss functions for solving the plot extraction problem. Further, we introduce APEX-1M - a new large scale dataset that contains both the plot images and the raw data. We demonstrate the performance of APEX-Net on the APEX-1M test set and show that it obtains impressive accuracy. We also show visual results of our network on unseen plot images and demonstrate that it extracts the shape of the plots to a great extent. Finally, we develop a GUI for plot extraction that can benefit the community at large. The dataset and code will be made publicly available.