Invisible Stitch: Generating Smooth 3D Scenes with Depth Inpainting

arXiv:2404.19758v1 Announce Type: new
Abstract: 3D scene generation has quickly become a challenging new research direction, fueled by consistent improvements of 2D generative diffusion models. Most prior work in this area generates scenes by iteratively stitching newly generated frames with existing geometry. These works often depend on pre-trained monocular depth estimators to lift the generated images into 3D, fusing them with the existing scene representation. These approaches are then often evaluated via a text metric, measuring the similarity between the …

3d scene generation 3d scenes abstract arxiv become consistent cs.cv diffusion diffusion models generated generative geometry improvements inpainting prior research stitching type work