VirtualModel: Generating Object-ID-retentive Human-object Interaction Image by Diffusion Model for E-commerce Marketing

arXiv:2405.09985v1 Announce Type: new
Abstract: Due to the significant advances in large-scale text-to-image generation by diffusion model (DM), controllable human image generation has been attracting much attention recently. Existing works, such as Controlnet [36], T2I-adapter [20] and HumanSD [10] have demonstrated good abilities in generating human images based on pose conditions, they still fail to meet the requirements of real e-commerce scenarios. These include (1) the interaction between the shown product and human should be considered, (2) human parts like …

abstract adapter advances arxiv attention commerce controlnet cs.cv diffusion diffusion model e-commerce good human human images image image generation images marketing object scale text text-to-image type