AVSplat: assist-view preconditioning fixes dense-view degradation in feed-forward 3DGS
zhenjun_zhao · x · 2026-09-11
AVSplat tackles a counterintuitive failure of pose-free feed-forward 3D Gaussian Splatting: more input views often degrade results as global attention dilutes and naive voxel fusion over-smooths. Each view first interacts lightly with a small set of relevance/diversity-chosen Assist Views, whose cached features stabilize correspondence; an adaptive temperature-aware voxel fusion sharpens attribution using occupancy and point confidence. Ablations show assist-view preconditioning prevents dense-view degradation, and AVSplat restores positive view scaling for novel view synthesis from uncalibrated multi-view images.
More from Research
- NVIDIA open-sources SoL-Pi: auto-research loops cut agent token cost by up to 54% — udmrzn · 2026-09-11
- Survey Reviews Overparameterized ML Theory Behind the Fall of Bias-Variance Tradeoff — udmrzn · 2026-09-11
- OpenAI Paper Constructs Finite-Time Blowup for Navier–Stokes, Verified in Lean by GPT-6 Astra — burny_tech · 2026-09-11
- TDDN Fuses DINOv3 with CleanDIFT, More Than Triples CLIP's Dense Prediction Accuracy — udmrzn · 2026-09-11
- Scientist wireheads a fly with boosted dopamine and forces it to doomscroll 'FlyTok' — DynamicWebPaige · 2026-09-11
- NeurIPS 2026 asks all authors to pick preferred presentation sites by Sept 30 — NeurIPSConf · 2026-09-11