Trust3R: ICML 2026 Paper Adds Evidential Uncertainty to Feed-Forward 3D Reconstruction

rsasaki0109 · x · 2026-09-12

• The author introduces Trust3R, a trust-aware feed-forward 3D reconstruction framework accepted to ICML 2026.
• Pain point: geometric foundation models like DUSt3R and MASt3R predict dense pointmaps from uncalibrated images in one forward pass, but their per-pixel confidence is heuristic — no probabilistic grounding, and downstream alignment, fusion, and SLAM modules get no calibrated signal.
• Method: pair a lightweight gated residual mean refinement with evidential learning under a Normal-Inverse-Wishart (NIW) prior; marginalization yields a closed-form multivariate Student-t predictive distribution, giving probabilistically grounded pointmap uncertainty with moderate inference overhead.

Original post →

More from Embodied

• Indie dev prototypes custom secure hardware in a day with Rust firmware and root-key isolation — wavefnx · 2026-10-31
• UCLA's ChairNav teaches autonomous wheelchairs to learn each rider's preferences — zhoubolei · 2026-10-31
• Amazon Echo kills local voice storage: everything goes to the cloud from March 28 — alex_verem · 2026-10-31
• HP paid $116M for Humane, then bricked every $699 AI Pin 10 days later — alex_verem · 2026-10-31
• Moxie, the $800 AI robot friend for kids, went silent overnight after shutdown — alex_verem · 2026-10-31
• Subvocal's YC-backed glasses decode silent speech at 200+ wpm, built in 4 months — garrytan · 2026-10-31

Embodied channel →