BLT: Bidirectional Layout Transformer for Controllable Layout Generation

Abstract

Creating visual layouts is an important step in graphic design. Automatic generation of such layouts is important as we seek scale-able and diverse visual designs. Prior works on automatic layout generation focus on unconditional generation, in which the models generate layouts while neglecting user needs for specific problems. To advance conditional layout generation, we introduce BLT, a bidirectional layout transformer. BLT differs from autoregressive decoding as it first generates a draft layout that satisfies the user inputs and then refines the layout iteratively.We verify the proposed model on multiple benchmarks with various fidelity metrics. Our results demonstrate two key advances to the state-of-the-art layout transformer models. First, our model empowers layout transformers to fulfill controllable layout generation. Second, our model slashes the linear inference time in autoregressive decoding into a constant complexity, thereby achieving 4x-10x speedups in generating a layout at inference time.* Work done during their research internship at Google. (a) Conditional (top) and unconditional (bottom) generation.

Процес Прийняття Рішень При Дизайні Ілюстрацій Настільних Ігор З Використанням Додатків Генеративного Штучного Інтелекту

“…The rapid development of technologies based on use of artificial intelligence, in particular, the generation of images by text (text-toimage), is changing the way we look at all approaches to design. This applies to design of buildings [1], clothing [2,3], web design [4], game design [5][6][7], as well as the design of multimedia or printed publications [8]. A large number of similar studies even allowed to single out a separate branch -generative artificial intelligence technologies (Generative AI -GAI).…”