GRCNN: Graph Recognition Convolutional Neural Network for Synthesizing Programs from Flow Charts

TL;DR

This paper introduces GRCNN, a neural network that recognizes flow chart structures from images to facilitate automatic program synthesis, achieving high accuracy and efficiency.

Contribution

The paper presents GRCNN, a novel end-to-end neural network for recognizing graph structures from images to improve program synthesis from flow charts.

Findings

Program synthesis accuracy is 66.4%.

Edge recognition accuracy is 94.1%.

Node recognition accuracy is 67.9%.

Abstract

Program synthesis is the task to automatically generate programs based on user specification. In this paper, we present a framework that synthesizes programs from flow charts that serve as accurate and intuitive specifications. In order doing so, we propose a deep neural network called GRCNN that recognizes graph structure from its image. GRCNN is trained end-to-end, which can predict edge and node information of the flow chart simultaneously. Experiments show that the accuracy rate to synthesize a program is 66.4%, and the accuracy rates to recognize edge and nodes are 94.1% and 67.9%, respectively. On average, it takes about 60 milliseconds to synthesize a program.