TriFlow trains a rectified-flow transformer directly on triplane latents extracted by the pretrained InstantMesh reconstruction model, using standardization, empty-space masking, and a perceptual rendering loss, and reports improved text-to-3D generation.
pi-gan: Periodic implicit generative adversarial networks for 3d-aware image synthesis
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Taming Feed-forward Reconstruction Models as Latent Encoders for 3D Generative Models
TriFlow trains a rectified-flow transformer directly on triplane latents extracted by the pretrained InstantMesh reconstruction model, using standardization, empty-space masking, and a perceptual rendering loss, and reports improved text-to-3D generation.