In this work we combine two research threads from Vision/ Graphics and Natural Language Processing to formulate an image generation task conditioned on attributes in a multi-turn setting. By multiturn, we mean the image is generated in a series of steps of user-specified conditioning information. Our proposed approach is practically useful and offers insights into neural interpretability. We introduce a framework that includes a novel training algorithm as well as model improvements built for the multi-turn setting. We demonstrate that this framework generates a sequence of images that match the given conditioning information and that this task is useful for more detailed benchmarking and analysis of conditional image generation methods.
@article{arxiv.1806.06183,
title = {The Neural Painter: Multi-Turn Image Generation},
author = {Ryan Y. Benmalek and Claire Cardie and Serge Belongie and Xiadong He and Jianfeng Gao},
journal= {arXiv preprint arXiv:1806.06183},
year = {2018}
}