将图像生成与重建分离，提升少步生成效率与质量。
Efficient Image Synthesis with Sphere Latent Encoder
- 用预训练编码器和独立球面潜空间去噪模型解耦架构。
- 在Animal-Faces等数据集上生成质量更高、速度更快。
- 适合追求高效高质图像生成的开发者或研究者。
少步图像生成发展迅速，一致性与均流方法显著减少采样步骤。尽管推理成本低，但这些方法常面临训练不稳定与可扩展性差的问题。Sphere Encoder 是一种新方案，可在少数步骤内生成高质量图像；然而其推理过程中需反复在像素空间与潜空间间转换，并在单一架构中联合优化重建与生成，导致计算效率低下且目标冲突。为此，本文将框架解耦为固定预训练图像编码器与独立的球面潜空间去噪模型。该方法在训练与推理中均避免重复像素空间操作，提升效率，并使重建与生成能独立优化。在 Animal-Faces、Oxford-Flowers 与 ImageNet-1K 数据集上，本方法在生成质量与推理速度上均显著优于 Sphere Encoder，且性能媲美强效少步与多步基线。
原文摘要 · Abstract (English)
Few-step image generation has seen rapid progress, with consistency and meanflow-based methods significantly reducing the number of sampling steps. Despite their low inference cost, these approaches often suffer from training instability and limited scalability. Sphere Encoder is a recent alternative that produces high-quality images in only a few steps; however, it requires repeated transitions between the pixel space and latent space during inference while jointly optimizing reconstruction and generation within a single architecture. This design leads to computational inefficiency and objective conflict between reconstruction and generation. To address these limitations, we decouple the framework into a fixed pretrained image encoder and a separate latent denoising model trained entirely in a spherical latent space. Our approach eliminates repeated pixel-space operations during training and inference, improving efficiency and allowing reconstruction and generation to specialize independently. On Animal-Faces, Oxford-Flowers and ImageNet-1K datasets, our method significantly outperforms Sphere Encoder in both generation quality and inference speed, while achieving competitive results against strong few-step and multi-step baselines.
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。