This work proposes Easy-3Dfy, a compositional workflow that decouples object generation from scene layout estimation, allowing specialized models to fully exploit their respective strengths, and demonstrates that explicit geometric composition can effectively combine high-quality object generation with strong spatial understanding.
Abstract
Single-image 3D scene construction is a fundamental problem in 3D vision. Recent methods such as SAM-3Dfy enable end-to-end reconstruction of 3D assets and their scene layouts, but still struggle with fine-grained scene geometry and asset quality. Meanwhile, specialized 3D object generators provide detailed object geometry, while 3D foundation models offer strong and generalizable spatial understanding. We propose Easy-3Dfy, a compositional workflow that decouples object generation from scene layout estimation, allowing specialized models to fully exploit their respective strengths. An RPF-based aligner then bridges these capabilities by establishing geometric correspondence between object-local meshes and partial scene pointmaps, recovering object layout for scene composition. Without requiring complete scene-level supervision, the proposed workflow consistently outperforms SAM-3Dfy across multiple geometric metrics on Scene-Maker and 3D-FUTURE200. Our results demonstrate that explicit geometric composition can effectively combine high-quality object generation with strong spatial understanding, while enabling the workflow to benefit independently from advances in both 3D object generation and foundation models.
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...
Hong-Ye Hou, Shuai Zhang, Hong-Jie Li et al.· 0 citations
Fysiverse-3D-Vision is proposed, a unified vision-language-geometry framework for generative 3D scene reconstruction and executable asset construction from a single image that establishes a shared representation where spatial reasoning and geometric reconstruction mutually enhance each other, allowing object layouts to...
SPOON is introduced, a framework that reformulates multi-view compositional 3D generation as scene-level, geometry-grounded pose reasoning and progressively organizes object poses and camera configurations into a coherent scene-level spatial arrangement.
Gui-Biao Liao, M. Xiang, Heng Li et al.· arXiv.org· 0 citations
Interactive 3D scenes serve as the foundational infrastructure for embodied AI, digital twins, and immersive metaverses, providing the essential physical and semantic substrate for realistic agent-environment interaction. While recent advances have made single-image 3D reconstruction increasingly feasible, existing met...
Xiao-Cong Zeng, Zehuan Huang, Yi-Ran Qin et al.· 0 citations
We propose Tetris3D, a generative framework for single-image 3D scene reconstruction that recovers objects which are physically and geometrically coherent as a scene. Existing methods often generate objects independently or couple them implicitly, providing limited guidance for ensuring fine-grained spatial compatibili...
Jaeyeon Kim, Ji-Ze Jang, Jongmin Lee et al.· 0 citations
Procedural materials, typically represented as computational graphs, are widely adopted in 3D content creation because they enable editable, reusable, and highly controllable appearance design. To automate this expertise-intensive and time-consuming process, we study the problem of reference-guided stylized procedural...
Peng-Fei Zhang, Yun-Fan Ye, Wen-Zhao Dai et al.· Proceedings of the 34th ACM...· 0 citations
Known for his clear and elegant writing style, Bertsekas shaped fields from control and optimization to large-scale computation and artificial intelligence.