Sidun Liu

Instead of decoding free latent tokens or reconstructing fully explicit 3D primitives, IRIS represents the scene as a latent neural field and renders novel views by querying this field under self-predicted cameras, preserving the flexibility and optimization stability of implicit modeling, while introducing stronger geometric structure than unconstrained latent rendering.

Spatially grounded tokens in LocusGS lead to coherent Gaussian distributions that dramatically improve rendering quality in 3D scene synthesis.