NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis

Pretrained weights for NAMVIS (NeurIPS 2026), a diffusion-free, geometry-conditioned next-scale autoregressive model for sparse-view novel view synthesis.

• 📄 Paper: · OpenReview
• 🌐 Project page: https://smileyenot983.github.io/namvis
• 💻 Code: https://github.com/smileyenot983/namvis

Files

File	Description
namvis_1b.pth	NAMVIS 1B transformer (256×256)
infinity_vae_d32reg.pth	Multi-scale VQ-VAE from Infinity (frozen, unchanged, MIT)

Details

• Resolution: 256×256
• Input: one or more posed source images + target camera poses
• Trained on a filtered subset (~200K objects) of Objaverse; see Objaverse for per-asset licenses

Citation

@inproceedings{khafizov2026namvis,
  title     = {{NAMVIS}: Next-Scale Autoregressive Multi-View Image Synthesis},
  author    = {Khafizov, Ramil and Statsenko, Ilya and Rakhimov, Ruslan and Komarichev, Artem and Wonka, Peter and Burnaev, Evgeny},
  booktitle = {Advances in Neural Information Processing Systems},
  year      = {2026}
}