This repository is part of our work in which we seek to learn to identify images of charts as they are posted on social media. We are adapting the VGGNet architecture proposed by Simonyan and Zisserman to the requirements of our chart identification problem. We train our system on a dataset consisting of chart images from the ReVision corpus and general-purpose images from ILSVRC-2012.
You can use our pre-trained model to identify whether an image on the web displays a data visualisation or not. In case it does, our model will try to predict the depicted chart type.
This project is licensed under the terms of the Apache 2.0 License.