Daichi Horita

Content-aware graphic layout generation aims to automatically arrange visual elements along with a given content, such as an e-commerce product image. In this paper, we argue that the current layout generation approaches suffer from the limited training data for the high-dimensional layout structure. We show that a simple retrieval augmentation can significantly improve the generation quality. Our model, which is named Retrieval-Augmented Layout Transformer (RALF), retrieves nearest neighbor layout examples based on an input image and feeds these results into an autoregressive generator. Our model can apply retrieval augmentation to various controllable generation tasks and yield high-quality layouts within a unified architecture. Our extensive experiments show that RALF successfully generates content-aware layouts in both constrained and unconstrained settings and significantly outperforms the baselines.

Image completion techniques have made significant progress in filling missing regions (i.e., holes) in images. However, large-hole completion remains challenging due to limited structural information. In this paper, we address this problem by integrating explicit structural guidance into diffusion-based image completion, forming our structure-guided diffusion model (SGDM). It consists of two cascaded diffusion probabilistic models: structure and texture generators. The structure generator generates an edge image representing plausible structures within the holes, which is then used for guiding the texture generation process. To train both generators jointly, we devise a novel strategy that leverages optimal Bayesian denoising, which denoises the output of the structure generator in a single step and thus allows backpropagation. Our diffusion-based approach enables a diversity of plausible completions, while the editable edges allow for editing parts of an image. Our experiments on natural scene (Places) and face (CelebA-HQ) datasets demonstrate that our method achieves a superior or comparable visual quality compared to state-of-the-art approaches. The code is available for research purposes at https://github.com/UdonDa/Structure_Guided_Diffusion_Model.

Vision-induced gustatory manipulation interfaces can help people with dietary restrictions feel as if they are eating what they want by modulating the appearance of the alternative foods they are eating in reality. However, it is still unclear whether vision-induced gustatory change persists beyond a single bite, how the sensation changes over time, and how it varies among individuals from different cultural backgrounds. The present paper reports on a user study conducted to answer these questions using a generative adversarial network (GAN)-based real-time image-to-image translation system. In the user study, 16 participants were presented somen noodles or steamed rice through a video see-through head mounted display (HMD) both in two conditions; without or with visual modulation (somen noodles and steamed rice were translated into ramen noodles and curry and rice, respectively), and brought food to the mouth and tasted it five times with an interval of two minutes. The results of the experiments revealed that vision-induced gustatory manipulation is persistent in many participants. Their persistent gustatory changes are divided into three groups: those in which the intensity of the gustatory change gradually increased, those in which it gradually decreased, and those in which it did not fluctuate, each with about the same number of participants. Although the generalizability is limited due to the small population, it was also found that non-Japanese and male participants tended to perceive stronger gustatory manipulation compared to Japanese and female participants. We believe that our study deepens our understanding and insight into vision-induced gustatory manipulation and encourages further investigation.