Get Started
Home
Topics
Search
Library
Research questionHow can multi-view 3D reconstruction predict precise depth while preserving controllable uncertainty over plausible shapes?Multi-view systems must infer surface locations along rays, even when image evidence supports several plausible 3D completions. The representation must therefore support precise depth estimates alongside varied, controllable reconstructions.
AI
Computer Vision
Diffusion Models
Image & Video Processing
Machine Learning
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.Learning Spherical Occupancy Profiles for Multi-View 3D Reconstruction and GenerationThe evidence concerns spherical occupancy profiles distilled from multi-view 3D Gaussian reconstructions. Experiments use 999 Google Scanned Objects with 48 turntable views, an independent 90-object test split, and real-photo validation on two DTU scenes; they cover discriminative depth prediction, unconditional sampling, image-conditioned multi-solution reconstruction, profile morphology, and transfer, without establishing broader generalization.research paper · Sep 4, 2026
Related questions
How can monocular 3D reconstruction generalize across unseen object categories and arbitrary viewpoints?How can we complete unseen 3D scene regions from sparse, unconstrained views without dense 3D supervision?How can self-supervised surround-view depth estimation remain geometrically consistent when camera views minimally overlap?How can pretrained 3D scene representations infer depth from unseen viewpoints without task-specific reconstruction training?