FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Views
September 16, 2026
FAMOS uses a Multi-state Articulation Transformer to predict movable-part segmentation and joint parameters from sparse, unordered partial point clouds. The model aggregates articulation cues across multiple observations to enable feed-forward 3D modeling.
HOW THIS AFFECTS YOU
●
builderThis enables more robust 3D object modeling from limited monocular or sparse sensor data.
●
researcherThe multi-state attention mechanism allows for effective reasoning over variable numbers of sparse observations.