D-Prism: Differentiable Primitives for Structured Dynamic Modeling
For computer vision and graphics, D-Prism addresses the lack of structured dynamic modeling by combining primitives and 3DGS, offering a new way to jointly capture geometry and articulated motion.
D-Prism extends differentiable primitives to dynamic scenes, enabling high-fidelity structured dynamic modeling of multi-part assemblies by binding 3DGS to primitive surfaces with a deformation network and adaptive control. It achieves precise geometry and motion tracking, outperforming existing methods on structured dynamic benchmarks.
Capturing both geometry and rigid motion for structured dynamic objects, like multi-part assemblies or jointed mechanisms, remains a key challenge. Existing dynamic methods, such as deformable meshes or 3DGS, rely on unstructured representations and fail to jointly model suitable geometry and articulated motion. Primitive-based methods excel at structured static scenes, but their dynamic potential is still unexplored. We propose D-Prism, the first framework to achieve high-fidelity structured dynamic modeling by extending differentiable primitives to the dynamic domain. Specifically, we bind 3DGS to primitive surfaces, leveraging their respective strengths in appearance and geometry. We introduce a deformation network to control primitive motion, ensuring it accurately matches the object's movement. Furthermore, we design a novel adaptive control strategy to dynamically adjust primitive counts, better matching objects' true spatial footprint. Experiments confirm that our method excels at structured dynamic modeling, providing both structured geometry and precise motion tracking.