Sculpt4D：基于稀疏注意力扩散 Transformer 生成 4D 形状
Sculpt4D: Generating 4D Shapes via Sparse-Attention Diffusion Transformers
把预训练 3D 扩散 Transformer 通过块稀疏注意力扩展成时序 4D 生成器，其时间衰减稀疏掩码锚定首帧保身份的思路可直接迁移到视频/动态 NeRF 等需要长程时序一致性的生成任务，并显著降低注意力开销。
原文依据 a Block Sparse Attention mechanism that preserves object identity by anchoring to the initial frame
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期三维生成建模的突破在静态形状合成上取得了显著进展，但真正意义上的动态 4D 生成仍难以实现，受制于时序伪影与高昂的计算需求。本文提出 Sculpt4D，一种原生 4D 生成框架，它将高效的时序建模无缝集成到预训练的三维扩散 Transformer（Hunyuan3D 2.1）中，从而缓解 4D 训练数据稀缺的问题。其核心是一种块稀疏注意力机制，通过锚定初始帧来保持物体身份，同时借助时间衰减的稀疏掩码捕捉丰富的运动动态。该设计以高保真度忠实建模复杂的时空依赖关系，同时规避了全注意力的二次开销，将网络总计算量降低 56%。由此，Sculpt4D 在时序一致的 4D 合成上确立了新的最优水平，为高效且可扩展的 4D 生成开辟了道路。
英文原文摘要 982 字符
Recent breakthroughs in 3D generative modeling have yielded remarkable progress in static shape synthesis, yet truly dynamic 4D generation remains elusive, hindered by temporal artifacts and prohibitive computational demand. We present Sculpt4D, a native 4D generative framework that seamlessly integrates efficient temporal modeling into a pretrained 3D Diffusion Transformer (Hunyuan3D 2.1), thereby mitigating the scarcity of 4D training data. At its core lies a Block Sparse Attention mechanism that preserves object identity by anchoring to the initial frame while capturing rich motion dynamics via a time-decaying sparse mask. This design faithfully models complex spatiotemporal dependencies with high fidelity, while sidestepping the quadratic overhead of full attention and reducing network total computation by 56%. Consequently, Sculpt4D establishes a new state-of-the-art in temporally coherent 4D synthesis, charting a path toward efficient and scalable 4D generation.