Abstract
We study the problem of inferring an object-centric scene representation from a single image, aiming to derive a representation that explains the image formation process, captures the scene's 3D nature, and is learned without supervision. Most existing methods on scene decomposition lack one or more of these characteristics, due to the fundamental challenge in integrating the complex 3D-to-2D image formation process into powerful inference schemes like deep networks. In this paper, we propose unsupervised discovery of Object Radiance Fields (uORF), integrating recent progresses in neural 3D scene representations and rendering with deep inference networks for unsupervised 3D scene decomposition. Trained on multi-view RGB images without annotations, uORF learns to decompose complex scenes with diverse, textured background from a single image. We show that uORF performs well on unsupervised 3D scene segmentation, novel view synthesis, and scene editing on three datasets.
Video
Unsupervised 3D Segmentation
Input view
Novel view
GT color image
GT segmentation
Slot Attention
w/o background
w/o prog. trn.
uORF (ours)
Novel View Synthesis
Input view
Novel view
Ground truth
NeRF-AE
w/o background
w/o prog. trn.
uORF (ours)
Scene Editing
I. Move obj.
Input image
Input view
Novel view
Ground truth
Slot Attention
NeRF-AE
w/o bg.
w/o prog. trn.
uORF (ours)
II. Change bg.
Bg. image (input image is same as above)
Input view
Novel view
Ground truth
Slot Attention
NeRF-AE
w/o bg.
w/o prog. trn.
uORF (ours)
BibTeX