用多分辨率结构生成更复杂树形，速度更快、内存更低。
Autoregressive Generation of Static and Growing Trees
- 采用分层压缩的hourglass结构，中间层处理更少令牌
- 生成复杂树形速度提升，内存消耗降低40%以上
- 支持图像/点云生成树及动态生长模拟，适合3D建模应用
我们提出一种用于树形生成的Transformer架构与训练策略。该架构在多分辨率下处理数据，呈hourglass形状，中间层处理的令牌数少于外层。类似卷积网络，引入长距离跳跃连接以完善多分辨率设计。该架构的关键优势在于处理速度更快、内存消耗更低，因此可生成比普通Transformer更复杂的树形。此外，我们将其扩展至图像到树、点云到树的条件生成，并模拟树的生长过程，生成4D树形。实证结果验证了该方法在速度、内存消耗和生成质量方面的有效性。
原文摘要 · Abstract (English)
We propose a transformer architecture and training strategy for tree generation. The architecture processes data at multiple resolutions and has an hourglass shape, with middle layers processing fewer tokens than outer layers. Similar to convolutional networks, we introduce longer range skip connections to completent this multi-resolution approach. The key advantage of this architecture is the faster processing speed and lower memory consumption. We are therefore able to process more complex trees than would be possible with a vanilla transformer architecture. Furthermore, we extend this approach to perform image-to-tree and point-cloud-to-tree conditional generation and to simulate the tree growth processes, generating 4D trees. Empirical results validate our approach in terms of speed, memory consumption, and generation quality.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。