Attributing Image Generative Models using Latent Fingerprints
Attributing Image Generative Models using Latent Fingerprints
复制标题
DOI：
10.48550/arxiv.2304.09752
复制
发表时间：
2023-04
期刊：
影响因子：
--
通讯作者：
Guangyu Nie;C. Kim;Yezhou Yang;Yi Ren
中科院分区：
文献类型：
--
作者：
Guangyu Nie;C. Kim;Yezhou Yang;Yi Ren
生成模型能够创建与自然内容没有区别的内容。此类模型的开源开发引起了人们对其滥用于恶意目的的风险的担忧。一种潜在的风险缓解策略是通过指纹识别生成模型。当前的指纹识别方法在稳健的归因准确性和生成质量之间表现出显着的权衡，但缺乏改进这种权衡的设计原则。本文研究了潜在语义维度作为指纹的使用，从中我们可以分析设计变量（包括指纹维度、强度和容量的选择）对准确性与质量权衡的影响。与之前的SOTA相比，我们的方法需要最少的计算量，并且更适用于大规模模型。我们使用 StyleGAN2 和潜在扩散模型来证明我们方法的有效性。
Generative models have enabled the creation of contents that are indistinguishable from those taken from nature. Open-source development of such models raised concerns about the risks of their misuse for malicious purposes. One potential risk mitigation strategy is to attribute generative models via fingerprinting. Current fingerprinting methods exhibit a significant tradeoff between robust attribution accuracy and generation quality while lacking design principles to improve this tradeoff. This paper investigates the use of latent semantic dimensions as fingerprints, from where we can analyze the effects of design variables, including the choice of fingerprinting dimensions, strength, and capacity, on the accuracy-quality tradeoff. Compared with previous SOTA, our method requires minimum computation and is more applicable to large-scale models. We use StyleGAN2 and the latent diffusion model to demonstrate the efficacy of our method.