SG-Splat: Pose-Free Feed-Forward 3DGS Decouples Geometry and Attributes for Stability
zhenjun_zhao · x · 2026-10-28
SG-Splat is a pose-free feed-forward 3D Gaussian Splatting framework that decouples geometry estimation from Gaussian attribute prediction and adds a coarse-to-fine refinement pipeline: Flow-Guided Pose Refinement (differentiable bundle adjustment on flow features) followed by Residual-Guided Geometry Correction using depth residuals. On DL3DV across 12–32 input views in both posed and pose-free settings, it beats MVSplat, DepthSplat, YoNoSplat, AnySplat and WorldMirror on PSNR/SSIM/LPIPS, with sharper, more artifact-free novel views.
More from Research
- Rebuilding 'Democracy Does Cause Growth': long-run GDP effect rises to 33.6% with 2024 data — ChenhaoTan · 2026-10-28
- YC-backed Wondersearch claims OpenAI-level retrieval on FiQA with no embeddings, no vector DB — antoine_chaffin · 2026-10-28
- AAAI/ICLR Reviewer: About Half of Reviews Look AI-Generated, Peer Review in Crisis — dsmirza · 2026-10-28
- A bestiary of chaotic attractors from the Field–Golubitsky map, each with Dₙ symmetry — S_Conradi · 2026-10-28
- Manifold AI's WorldScape R²SI Tops RoboDojo Agent Track by Turning Robot Failures Into Skills — jiqizhixin · 2026-10-28
- AXA Uses NVIDIA Earth-2 to Simulate 1,000+ Cyclone Seasons for Storm Risk Modeling — NVIDIAAI · 2026-10-28