G6 Radiance Field Foundations
- K2E-B-G6-33D Gaussian SplattingMar 31, 2024
3DGS 论文笔记 — 各向异性 3D 高斯显式表示 + 可微 tile 光栅化,1080p ≥ 30 FPS 实时渲染,质量达 SOTA;3DGS-SLAM 系列的基石
- K2E-B-G6-2Instant-NGPMar 31, 2024
Instant-NGP 论文笔记 — 多分辨率哈希编码 + 小 MLP,把 NeRF 训练从数小时压到秒级(NeRF 任务 15 s 起、1–5 min 达 mip-NeRF 质量),validated 四个任务;多个 NeRF-SLAM 的渲染后端
- K2E-B-G6-1NeRF — Neural Radiance FieldsMar 31, 2024
NeRF 论文笔记 — 5D MLP 表示连续辐射场,可微体渲染积分、位置编码、分层采样、定量结果与代码走读。NeRF-SLAM 系列的表示基石
- K2E-B-G6-4NerfiesMar 31, 2024
Nerfies 论文笔记 — 可形变 NeRF,per-frame SE(3) 形变场将观测坐标 warp 到 canonical 模板,弹性正则 + coarse-to-fine 位置编码,从手机自拍视频重建可形变场景
- K2E-B-G6-5SMERFMar 31, 2024
SMERF 论文笔记 — 可流式省内存辐射场,K³ 子模型网格 + Zip-NeRF 蒸馏,普通设备实时漫游大场景 (300 m²,~200 FPS)
- K2E-B-G6-6These Magic Moments: Differentiable Uncertainty Quantification of Radiance Field ModelsJul 1, 2026
Magic Moments paper note — reinterprets volume rendering as a stochastic process to derive closed-form higher-order moments (variance) without extra training; applies to both NeRF and 3DGS (2 ms/frame real-time UQ on 3DGS), covering color/depth/semantics
- K2E-B-G6-73DGS-U: Predictive Photometric Uncertainty in Gaussian SplattingJul 5, 2026
Paper note on 3DGS-U — post-hoc, plug-and-play uncertainty estimation for 3DGS via Bayesian-regularized linear least-squares over training residuals; view-dependent per-primitive uncertainty channel rendered at the same speed as RGB