Abstract
Scaling image-to-3D generation to ultra-high resolutions requires controlling rapidly growing computational costs without sacrificing fine geometric detail. We present \textbf{Filigree3D}, a sparse latent flow-matching framework that generates 3D geometry from a single image at voxel resolutions up to $2048^3$, with straightforward extensibility to $4096^3$. To make training tractable, we introduce Structure-Aware Sparse Scaling, which combines spatial bounding with alternating local-global attention to constrain token growth while preserving both fine-scale details and long-range structural context. To enhance detail reconstruction, we curate training samples based on their high-resolution geometric gains and inject multi-scale image features into a sparse 3D DiT, effectively coupling structural semantics with fine-grained visual cues. Furthermore, a visibility-aware voxel regularization strategy improves robustness against sparse perturbations and facilitates the completion of unobserved geometry. Under our default configuration, Filigree3D maintains peak GPU memory consumption within practical limits for contemporary hardware, enabling the generation of highly intricate 3D geometry in approximately one minute. Extensive experiments demonstrate that our method yields substantial improvements in overall geometric fidelity and fine-detail preservation compared to existing baselines, validating practical, detail-preserving 3D generation at unprecedented resolutions.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Li, H., Yang, X., Wu, X., Lyu, J., & lv, C. (2026). FILIGREE3D: Scaling Sparse Latent Flow Matching for Ultra-High-Resolution Image-to-3D Generation. https://omanscience.com/en/articles/filigree3d-scaling-sparse-latent-flow-matching-for-ultra-high-resolution-image-to-3d-generation
MLA 9
Li, Hongjie, et al. "FILIGREE3D: Scaling Sparse Latent Flow Matching for Ultra-High-Resolution Image-to-3D Generation." https://omanscience.com/en/articles/filigree3d-scaling-sparse-latent-flow-matching-for-ultra-high-resolution-image-to-3d-generation.
Chicago (author–date)
Li, Hongjie, Xinran Yang, Xiuchao Wu, Jiangjing Lyu, and chengfei lv. 2026. "FILIGREE3D: Scaling Sparse Latent Flow Matching for Ultra-High-Resolution Image-to-3D Generation." https://omanscience.com/en/articles/filigree3d-scaling-sparse-latent-flow-matching-for-ultra-high-resolution-image-to-3d-generation.
Harvard
Li, H., Yang, X., Wu, X., Lyu, J. and lv, C. (2026) 'FILIGREE3D: Scaling Sparse Latent Flow Matching for Ultra-High-Resolution Image-to-3D Generation', Available at: https://omanscience.com/en/articles/filigree3d-scaling-sparse-latent-flow-matching-for-ultra-high-resolution-image-to-3d-generation.
Vancouver
Li H, Yang X, Wu X, Lyu J, lv C. FILIGREE3D: Scaling Sparse Latent Flow Matching for Ultra-High-Resolution Image-to-3D Generation. https://omanscience.com/en/articles/filigree3d-scaling-sparse-latent-flow-matching-for-ultra-high-resolution-image-to-3d-generation
IEEE
H. Li, X. Yang, X. Wu, J. Lyu, and C. lv, "FILIGREE3D: Scaling Sparse Latent Flow Matching for Ultra-High-Resolution Image-to-3D Generation," https://omanscience.com/en/articles/filigree3d-scaling-sparse-latent-flow-matching-for-ultra-high-resolution-image-to-3d-generation.