MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation
This addresses scalability issues for researchers and practitioners in 3D generation, though it is incremental as it builds on existing attention-based methods.
The paper tackles the problem of poor scalability in part-aware 3D generation due to quadratic attention costs by introducing MoCA, which uses importance-based component routing and compression to enable efficient compositional 3D asset creation, outperforming baselines in experiments.
Compositionality is critical for 3D object and scene generation, but existing part-aware 3D generation methods suffer from poor scalability due to quadratic global attention costs when increasing the number of components. In this work, we present MoCA, a compositional 3D generative model with two key designs: (1) importance-based component routing that selects top-k relevant components for sparse global attention, and (2) unimportant components compression that preserve contextual priors of unselected components while reducing computational complexity of global attention. With these designs, MoCA enables efficient, fine-grained compositional 3D asset creation with scalable number of components. Extensive experiments show MoCA outperforms baselines on both compositional object and scene generation tasks. Project page: https://lizhiqi49.github.io/MoCA