Get Started
Home
Topics
Search
Library
Research questionHow can sparse-view 3D scene representations avoid cross-view artifacts without costly per-scene reconstruction?With few input views, NeRF and 3D Gaussian Splatting can produce geometry and appearance that disagree across viewpoints. Existing artifact correction may require paired corrupted and clean renders plus costly reconstruction for each scene, limiting scalable training.
AI
Computer Vision
Diffusion Models
Image & Video Processing
Machine Learning
Research Paper
Robotics
Technology
Video Generation
Latest papersRecent research connected to this question, newest first.Rethinking 3D Noise: Learning 3D-Aware Video Priors via Optimization-Free Morphological PerturbationsThe source focuses on explicit 3D Gaussian Splatting and optimization-free morphological perturbations over Gaussian scale, rotation, and pruning. It reports diagnostic experiments in a lightweight video diffusion sandbox, scaling through ControlNet to a 14B-parameter video model, with downstream robotics results reported for four manipulation tasks.research paper · Sep 3, 2026
Related questions
How can gradient-based 3D Gaussian Splatting reconstruct static and dynamic scenes reliably when distant pixels provide weak updates?How can existing 3D Gaussian Splatting models be compacted for storage and transmission while preserving novel-view fidelity?How can feed-forward 3D Gaussian Splatting meet a fixed Gaussian budget without sacrificing novel-view quality?How can 3D Gaussian Splatting training stay efficient as high-resolution scenes require more Gaussian primitives?