@zhiwen_fan_: paper from dust3r’s team
Summary
A paper from the DUSt3R team proposes sparse auto-regressive modeling for 3D scene generation from multi-view images, using a voxel-aligned 3D latent space and an occupancy-aware masked autoregressive transformer.
View Cached Full Text
Cached at: 09/06/26, 10:48 AM
paper from dust3r’s team
Zhenjun Zhao (@zhenjun_zhao): Sparse auto-regressive modeling for scene generation from multi-view images
Thomas Lucas, Maxime Pietrantoni, @WeinzaepfelP, Wonjune Cho, @bardienus, @Vinc3nt_Leroy, @JeromeRevaud
tl;dr: voxel-aligned 3D latent space + occupancy-aware masked autoregressive transformer
Similar Articles
@vincieye: Generating full 3D scenes from just a few photos usually means hallucinating blind spots or blowing up memory. SPAR3S s…
SPAR3S is a sparse auto-regressive generative model that efficiently generates complete 3D scenes from sparse multi-view images without requiring 3D ground truth data, using a voxel-aligned latent space and differentiable 3D Gaussian Splatting.
GAE: Learning a Geometry-Native Latent Space for 3D-Consistent World Generation
The paper introduces GAE, a geometry-native autoencoder that creates a compact latent space for generating 3D-consistent scenes, enhancing visual quality and coherence over existing methods.
Native and Compact Structured Latents for 3D Generation
This paper introduces O-Voxel, a new sparse voxel representation for 3D generative modeling that efficiently handles complex topologies and appearance, and trains large-scale flow-matching models with 4B parameters to achieve state-of-the-art generation quality.
3D Masked Autoencoders are Robust Learners of Volumetric and Multimodal Cellular Representations for Microscopy
This paper presents 3D masked autoencoders for volumetric microscopy data, demonstrating that 3D modeling outperforms 2D max-projection and slice-based variants on downstream single-cell tasks, with cross-modal alignment to a protein language model further improving performance.
Beyond Pixels: From Video Priors to 4D Worlds
This paper introduces Latent-to-4D, a method for direct 4D scene generation from video diffusion latents without retraining across generators, achieving better geometry and temporal stability than cascaded approaches.