Argus improves indoor panoramic 3D reconstruction with covisibility and geometry transformers

ducha_aiki · x · 2026-07-21

Argus is a paper on metric panoramic 3D reconstruction for indoor scenes.

• The method introduces a covisibility transformer to pick the best reference frame before reconstruction.
• A geometry transformer then aggregates multi-view geometric features for downstream predictions such as camera pose, depth, camera points, rotated points, and world points.
• The authors say this reference-frame selection is important for indoor VGGT-style reconstruction.
• On RealSee3D, the paper reports better results than VGGT360, MapAnything360, and a c360 across alignment setups.
• Ablations show reference-view learning and the geometric losses materially improve performance.

Original post →

More from Embodied

• NVIDIA says physical AI starts in simulation with OpenUSD and synthetic data — MonaJalal_ · 2026-07-22
• Tesla expands Robotaxi rides to seven areas, including new Orlando and Tampa zones — elonmusk · 2026-07-22
• Hands-on robotics workshop on Saturday may be the last in-person session before August — StewartalsopIII · 2026-07-22
• NVIDIA pitches World Foundation Models as a way to scale physical AI data generation — MonaJalal_ · 2026-07-22
• RoboMME Podcast Preview: Benchmarking Memory for Robotic Policies — chris_j_paxton · 2026-07-21
• Gritt says an 8-person crew now installs 3,000 to 4,000 solar panels a day — HaktanSuren · 2026-07-21

Embodied channel →