3D-aware Image Generation using 2D Diffusion Models
This addresses the problem of generating 3D-consistent images from 2D data for applications in computer vision and graphics, representing a novel method for a known bottleneck rather than an incremental improvement.
The paper tackles 3D-aware image generation by leveraging 2D diffusion models, formulating it as multiview image set generation and using depth information from monocular estimators to train on ImageNet, resulting in high-quality images that significantly outperform prior methods and handle large view angles from unaligned real-world data.
In this paper, we introduce a novel 3D-aware image generation method that leverages 2D diffusion models. We formulate the 3D-aware image generation task as multiview 2D image set generation, and further to a sequential unconditional-conditional multiview image generation process. This allows us to utilize 2D diffusion models to boost the generative modeling power of the method. Additionally, we incorporate depth information from monocular depth estimators to construct the training data for the conditional diffusion model using only still images. We train our method on a large-scale dataset, i.e., ImageNet, which is not addressed by previous methods. It produces high-quality images that significantly outperform prior methods. Furthermore, our approach showcases its capability to generate instances with large view angles, even though the training images are diverse and unaligned, gathered from "in-the-wild" real-world environments.