Built on native brick tokenization and backed by our 55,000-model LegoVerse dataset, LegoACE is an autoregressive engine that generates expressive LEGO assemblies from text or normal maps.
TripoSplat introduces Density-Sampled Gaussians (DeG) for fully adaptive, grid-free 3D generation. We achieve differentiable densification by parameterizing primitive centers as samples from a learnable spatial density, optimized directly via a novel render-loss gradient. To model these unstructured sets, our VecSeq diffusion framework resolves permutation ambiguity by anchoring latents to a deterministic 3D Sobol sequence. TripoSplat achieves state-of-the-art single-image-to-3D generation while uniquely enabling variable-resolution decoding from a single compact latent code.
By jointly representing shape, skeleton, and skinning as mutually consistent S³ Fields, AniGen produces animation-ready 3D assets from a single image, substantially outperforming prior methods in rig validity.