REVIEW 4 cited by
DM-NeRF: 3D Scene Geometry Decomposition and Manipulation from 2D Images
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
In this paper, we study the problem of 3D scene geometry decomposition and manipulation from 2D views. By leveraging the recent implicit neural representation techniques, particularly the appealing neural radiance fields, we introduce an object field component to learn unique codes for all individual objects in 3D space only from 2D supervision. The key to this component is a series of carefully designed loss functions to enable every 3D point, especially in non-occupied space, to be effectively optimized even without 3D labels. In addition, we introduce an inverse query algorithm to freely manipulate any specified 3D object shape in the learned scene representation. Notably, our manipulation algorithm can explicitly tackle key issues such as object collisions and visual occlusions. Our method, called DM-NeRF, is among the first to simultaneously reconstruct, decompose, manipulate and render complex 3D scenes in a single pipeline. Extensive experiments on three datasets clearly show that our method can accurately decompose all 3D objects from 2D views, allowing any interested object to be freely manipulated in 3D space such as translation, rotation, size adjustment, and deformation.
Forward citations
Cited by 4 Pith papers
-
CF3: Compact and Fast 3D Feature Fields
CF3 builds a compact 3D feature field from a pre-trained 3DGS by feature lifting, per-Gaussian autoencoding, and adaptive sparsification, matching baseline segmentation quality with roughly 5% of the Gaussians.
-
Layered Motion Fusion: Lifting Motion Segmentation to 3D in Egocentric Videos
A layered neural radiance field fused with 2D motion masks and refined at test time beats both the 2D motion segmentation baseline and previous 3D methods on dynamic object segmentation in egocentric video.
-
Enhancing LLM Training via Spectral Clipping
SPECTRA improves LLM pretraining via post-clipping of update spectral norms and optional pre-clipping of gradient spikes, framed as Composite Frank-Wolfe regularization.
-
Leveraging 2D Priors and SDF Guidance for Dynamic Urban Scene Rendering
UGSDF achieves state-of-the-art novel-view rendering of dynamic urban objects without LiDAR or 3D motion annotations by jointly optimizing SDFs and 3D Gaussians under 2D depth and point-tracking priors.
Discussion (0). Continue with ORCID to comment.