The Neural Painter: Multi-Turn Image Generation
This work addresses the need for more detailed analysis and benchmarking in conditional image generation for researchers in vision and graphics, though it appears incremental as it combines existing research threads.
The paper tackles the problem of generating images through a multi-turn process where user-specified conditioning attributes guide the generation in steps, resulting in a framework that produces sequences of images matching the conditioning information and aids in benchmarking conditional image generation methods.
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.