PAGaS refines multi-view stereo depths by optimizing 1DoF Gaussians whose positions and sizes are fixed by back-projected pixel volumes, producing detailed depth maps that outperform reference baselines on 3D reconstruction benchmarks.
arXiv preprint arXiv:2401.11673 , year=
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.CV 4verdicts
UNVERDICTED 4roles
other 1polarities
unclear 1representative citing papers
GeoQuery replaces corrupted rendering features with geometry-aligned proxy queries and restricts cross-view attention to local windows, enabling robust diffusion-based refinement under extreme view sparsity.
Procedural rules with NURBS generate MVS training data that outperforms same-scale manual curation and matches or exceeds larger manual datasets.
A technique reconstructs large urban areas from sparse extreme off-nadir satellite images by modeling geometry as a Z-monotonic 2.5D height map SDF and applying a generative network to restore plausible textures on the resulting mesh.
citing papers explorer
-
PAGaS: Pixel-Aligned 1DoF Gaussian Splatting for Depth Refinement
PAGaS refines multi-view stereo depths by optimizing 1DoF Gaussians whose positions and sizes are fixed by back-projected pixel volumes, producing detailed depth maps that outperform reference baselines on 3D reconstruction benchmarks.
-
GeoQuery: Geometry-Query Diffusion for Sparse-View Reconstruction
GeoQuery replaces corrupted rendering features with geometry-aligned proxy queries and restricts cross-view attention to local windows, enabling robust diffusion-based refinement under extreme view sparsity.
-
SimpleProc: Fully Procedural Synthetic Data from Simple Rules for Multi-View Stereo
Procedural rules with NURBS generate MVS training data that outperforms same-scale manual curation and matches or exceeds larger manual datasets.
-
From Orbit to Ground: Generative City Photogrammetry from Extreme Off-Nadir Satellite Images
A technique reconstructs large urban areas from sparse extreme off-nadir satellite images by modeling geometry as a Z-monotonic 2.5D height map SDF and applying a generative network to restore plausible textures on the resulting mesh.