小波潜在扩散（WaLa）：具有紧凑小波编码的十亿参数3D生成模型
Wavelet Latent Diffusion (Wala): Billion-Parameter 3D Generative Model with Compact Wavelet Encodings
AI总结：
本文提出Wavelet Latent Diffusion（WaLa）方法，通过将3D形状编码为紧凑的小波潜在表示实现2427倍压缩，训练出约十亿参数的高效3D生成模型，在生成质量与计算效率上达到SOTA。
摘要：
大规模3D生成模型需要大量计算资源，但往往难以在高分辨率下捕捉精细细节和复杂几何结构。我们将此局限归因于当前表示方法的低效性，它们缺乏有效建模生成模型所需的紧凑性。为解决这一问题，我们引入了一种名为Wavelet Latent Diffusion（或WaLa）的新方法，将3D形状编码为基于小波的紧凑潜在编码。具体而言，我们将$256^3$的符号距离场压缩为$12^3 \ imes 4$的潜在网格，实现了高达2427倍的压缩比，且细节损失极小。这种高压缩度使我们的方法能在不增加推理时间的情况下高效训练大规模生成网络。我们的模型（包含条件和无条件模型）包含约十亿个参数，并成功生成了$256^3$分辨率的高质量3D形状。此外，尽管模型规模庞大，WaLa仍提供快速推理，根据条件不同可在2至4秒内生成形状。我们在多个数据集上展示了最先进的性能，在生成质量、多样性和计算效率方面均有显著提升。我们开源了代码，并据我们所知，发布了跨不同模态的最大预训练3D生成模型。
英文摘要：
Large-scale 3D generative models require substantial computational resources yet often fall short in capturing fine details and complex geometries at high resolutions. We attribute this limitation to the inefficiency of current representations, which lack the compactness required to model the generative models effectively. To address this, we introduce a novel approach called Wavelet Latent Diffusion, or WaLa, that encodes 3D shapes into wavelet-based, compact latent encodings. Specifically, we compress a $256^3$ signed distance field into a $12^3 \times 4$ latent grid, achieving an impressive 2427x compression ratio with minimal loss of detail. This high level of compression allows our method to efficiently train large-scale generative networks without increasing the inference time. Our models, both conditional and unconditional, contain approximately one billion parameters and successfully generate high-quality 3D shapes at $256^3$ resolution. Moreover, WaLa offers rapid inference, producing shapes within two to four seconds depending on the condition, despite the model's scale. We demonstrate state-of-the-art performance across multiple datasets, with significant improvements in generation quality, diversity, and computational efficiency. We open-source our code and, to the best of our knowledge, release the largest pretrained 3D generative models across different modalities.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
