Skip to main content
Search

# RDT-3D

1 related articles

Subscribe to updates
  1. arXiv

    3D generation trains and infers at mismatched resolutions: RDT-3D anchors guidance at 512 while supervising at 1536, lifting F-Score to 0.6886

    In two-stage high-fidelity 3D geometry generation, training aligns coordinate guidance and geometric supervision at one resolution, while inference decodes high-resolution geometry from coarse coordinates, and this mismatch over-smooths thin structures. RDT-3D fixes rotary position embedding (RoPE) spatial guidance at a coarse 512 resolution and keeps geometric supervision at 1536, forcing the refinement network to extrapolate detail from imprecise positional cues. On Toys4K, the RDT model reaches a Chamfer distance of 0.0164 and an F-Score of 0.6886, against 0.0184 and 0.6651 for the symmetric 1536 baseline.

You’re all caught up