Bri3D: Bridging the Object-Scene Gap in Single-Image 3D Generation via Scene-Anchor
cs.CV
Submitted: 2026-11-05
Updated: 2026-11-05
License: http://creativecommons.org/licenses/by/4.0/
The gist: Generating compositional 3D scenes from a single image requires recovering detailed object geometry and coherent spatial arrangements.
Terminology
Abstract
Generating compositional 3D scenes from a single image requires recovering detailed object geometry and coherent spatial arrangements. Existing object-to-scene methods, including SAM3D and MIDI, extend object generation through image-context reasoning or inter-instance modeling. Coordinating objects remains challenging with limited scene-level supervision. Directly extending object-centric generators to entire scenes can degrade geometry and fail to preserve instance scale, orientation, and pose. We propose Bri3D, a framework bridging scene-level structural reasoning and object-level generative priors through instance-aware voxel scene anchors. Bri3D predicts a color-coded scene scaffold encoding coarse object geometry and spatial arrangements. Each instance is extracted as a scene anchor and refined into a pose-preserving, prior-compatible voxel support, enabling a frozen object generator to synthesize detailed textured geometry in the anchor-defined spatial reference. This coarse-to-fine design separates scene organization from object-scale detail synthesis, focusing scene-level training on spatial organization and anchor refinement while reusing pretrained object priors. Experiments demonstrate improvements in scene layout, object geometry, visual quality, and generalization. Bri3D predicts scene layouts efficiently; full scene generation is faster than SceneGen and SAM3D.
Sources
- Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attention
- HiScene: Creating Hierarchical 3D Scenes with Isometric View Generation
- TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Models
- HoloRecon: Incremental Reconstruction of Long-Range Panoramas with Global Consistency
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers
- Flow Matching for Generative Modeling
- SceneGen: Single-Image 3D Scene Generation in One Feedforward Pass
- DINOv2: Learning Robust Visual Features without Supervision
- TripoSR: Fast 3D Object Reconstruction from a Single Image
- Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformer
- Direct3D-S2: Gigascale 3D Generation Made Easy with Spatial Sparse Attention
- CitySTAR: Structured and Topology-Aware Reasoning for Open-Vocabulary Urban 3D Grounding
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs