用一个参数控制生成细节的精细程度，无需重训练。
Omegance: A Single Parameter for Various Granularities in Diffusion-Based Synthesis
- 仅通过一个参数ω在去噪过程中调节生成细节粒度。
- 支持局部区域或时间步的精细化控制，效果显著。
- 适配多种扩散模型，适合需要灵活细节控制的场景。
本文表明，仅需一个参数ω即可有效控制基于扩散模型的合成中生成内容的粒度。该参数嵌入扩散模型反向过程的去噪步骤中，无需模型重训练或架构修改，计算开销极小，却能精确调控生成结果的细节水平。通过施加空间掩码或随时间变化的ω值，可实现区域或时间步级的粒度控制。外部控制信号或参考图像可用于生成精确的ω掩码，实现定向细节调整。尽管方法简单，其在多种图像与视频生成任务中均表现出色，且可适配先进扩散模型。代码已公开于 https://github.com/itsmag11/Omegance。
原文摘要 · Abstract (English)
In this work, we show that we only need a single parameter $ω$ to effectively control granularity in diffusion-based synthesis. This parameter is incorporated during the denoising steps of the diffusion model's reverse process. This simple approach does not require model retraining or architectural modifications and incurs negligible computational overhead, yet enables precise control over the level of details in the generated outputs. Moreover, spatial masks or denoising schedules with varying $ω$ values can be applied to achieve region-specific or timestep-specific granularity control. External control signals or reference images can guide the creation of precise $ω$ masks, allowing targeted granularity adjustments. Despite its simplicity, the method demonstrates impressive performance across various image and video synthesis tasks and is adaptable to advanced diffusion models. The code is available at https://github.com/itsmag11/Omegance.
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。