RIGOR: Treating 360° Panoramas as Virtual Camera Rigs for Omnidirectional Reconstruction
kwangmoo_yi · x · 2026-09-17
kwangmoyi released RIGOR (Rig-Informed Geometry for Omnidirectional Reconstruction) with code on GitHub.
- Each 360° panorama is split into four perspective views treated as a virtual rig with a shared optical center and known geometry; enforcing cross-view agreement lets existing feed-forward reconstruction models work with 360 camera input.
- The system performs long-sequence reconstruction from gravity-aligned panoramas with a frozen perspective backbone, integrating Grounded-SAM-2 and a Depth Anything 3 streaming module.
- The repo ships batch scripts, environment setup, and weight downloads for the Hilti-Trimble SLAM Challenge 2026, making evaluation reproducible.
Related event: ETH Zurich's RIGOR Tackles Omnidirectional Reconstruction Drift(3 posts)→
More from Research
- MIT camp builds self-healing 'Frankensteel' using shape-memory wires — TinfoilTricorn · 2026-09-17
- Allen Institute maps topographic structure and function of locus coeruleus noradrenaline neurons in Nature — adam_k_glaser · 2026-09-17
- Allen Institute releases 3D-MAESTRO, an automated pipeline for large-scale volumetric brain microscopy — adam_k_glaser · 2026-09-17
- Stanford's James Zou: language may be <1% of intelligence, so LLMs won't reach AGI — james_y_zou · 2026-09-17
- VC-Attention: training-free low-bit attention hits 1.9x on B200, beating FlashAttention-4 — xiuyu_l · 2026-09-17
- Boltz fused kernels deliver nearly 10x protein folding throughput gain — AllThingsApx · 2026-09-17