MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation

arXiv:2404.03656v1 Announce Type: new
Abstract: We present MVD-Fusion: a method for single-view 3D inference via generative modeling of multi-view-consistent RGB-D images. While recent methods pursuing 3D inference advocate learning novel-view generative models, these generations are not 3D-consistent and require a distillation process to generate a 3D output. We instead cast the task of 3D inference as directly generating mutually-consistent multiple views and build on the insight that additionally inferring depth can provide a mechanism for enforcing this consistency. Specifically, we …

abstract arxiv consistent cs.cv distillation fusion generate generative generative modeling generative models images inference modeling novel process rgb-d type via view