SPAR3S generates complete 3D scenes from sparse views via sparse voxel-aligned autoregressive modeling
zhenjun_zhao · x · 2026-09-05
A Naver Labs Europe paper introduces SPAR3S, a sparse voxel-aligned 3D latent generative model for conditional scene completion from unconstrained multi-view images, requiring no ground-truth 3D supervision. The latent space — representing only occupied voxels — is learned from images via photometric supervision through differentiable 3D Gaussian Splatting. A masked autoregressive transformer jointly predicts voxel occupancy and latent tokens, enabling spatially consistent generation of unseen regions, validated on synthetic indoor scenes.
Related event: SPAR3S Generates Complete 3D Scenes from Sparse Views(3 posts)→
More from Multimodal
- Qwen2.1 image gen sampler and scheduler combos benchmarked on locked seed — FreeTheClanks · 2026-09-22
- Runway expands Workflows with Compositing, Alpha, HDR, and Depth Map nodes — runwayml · 2026-09-22
- Grok-style character prompts packaged into a copy-paste web app — dejavucoder · 2026-09-22
- Hyper3D Agentic Mode turns 3D modeling into a conversation with editable, production-ready models — hey_abusiddik · 2026-09-22
- Tilt-Shift Videos With MiniMax H3: Full Local Workflow Shown Off — alisitskii · 2026-09-22
- Peter Diamandis showcases two AI-made short films, SLINGSHOT and THE GIFTED — PeterDiamandis · 2026-09-22