Customizing Text-to-Image Diffusion with Camera Viewpoint Control

arXiv:2404.12333v1 Announce Type: new
Abstract: Model customization introduces new concepts to existing text-to-image models, enabling the generation of the new concept in novel contexts. However, such methods lack accurate camera view control w.r.t the object, and users must resort to prompt engineering (e.g., adding "top-view") to achieve coarse view control. In this work, we introduce a new task -- enabling explicit control of camera viewpoint for model customization. This allows us to modify object properties amongst various background scenes via …

abstract arxiv concept concepts control cs.cv customization diffusion enabling engineering however image image diffusion model customization novel object prompt text text-to-image type view work