Table 3 Experiments comparing training and inference times for different models.
From: Generation driven understanding of localized 3D scenes with 3D diffusion model
Method | Training time | Inference time (seconds/sample) |
|---|---|---|
\(\pi\)-GAN | 242 hours | 0.69 |
EG3D | 296 hours | 1.31 |
DiffRF | 194 hours | 0.87 |
AutoSDF | 257 hours | 0.76 |
3D-UDDPM(Ours) | 239 hours | 0.84 |
