geometric priors
Prior knowledge or assumptions about the geometric properties of data distributions, which can guide model training and improve performance in tasks like shape recognition.
- AtlasGS: Atlanta-world Guided Surface Reconstruction with Implicit Structured Gaussians
- Attention on the Sphere
- Flux4D: Flow-based Unsupervised 4D Reconstruction
- HumanCrafter: Synergizing Generalizable Human Reconstruction and Semantic 3D Segmentation
- KaRF: Weakly-Supervised Kolmogorov-Arnold Networks-based Radiance Fields for Local Color Editing
- MS-GS: Multi-Appearance Sparse-View 3D Gaussian Splatting in the Wild
- PlanarGS: High-Fidelity Indoor 3D Gaussian Splatting Guided by Vision-Language Planar Priors
- Rectified Point Flow: Generic Point Cloud Pose Estimation
- Template-Guided 3D Molecular Pose Generation via Flow Matching and Differentiable Optimization