Attributing Image Generative Models using Latent Fingerprints
Authors: Guangyu Nie, Changhoon Kim, Yezhou Yang, Yi Ren
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments are conducted for both with and without a combination of postprocesses including (image noising, blurring, and JPEG compression, and their combination). We conduct experiments on SG2 (Karras et al., 2020) and LDM (Rombach et al., 2022) models trained on various datasets including FFHQ (Karras et al., 2019), AFHQ-Cat, and AFHQ-Dog (Choi et al., 2020). We present generation quality results in Table 1. |
| Researcher Affiliation | Academia | 1 School for Engineering of Matter, Transport and Energy, Arizona State University 2School of Computing and Augmented Intelligence, Arizona State University. Correspondence to: Yezhou Yang <yz.yang@asu.edu>, Yi Ren <yren32@asu.edu>. |
| Pseudocode | No | The paper provides mathematical formulations and propositions but does not include any pseudocode or explicitly labeled algorithm blocks. |
| Open Source Code | Yes | Codes are available in github. |
| Open Datasets | Yes | We conduct experiments on SG2 (Karras et al., 2020) and LDM (Rombach et al., 2022) models trained on various datasets including FFHQ (Karras et al., 2019), AFHQ-Cat, and AFHQ-Dog (Choi et al., 2020). |
| Dataset Splits | No | The paper states it uses models trained on various datasets (FFHQ, AFHQ-Cat, AFHQ-Dog) and specifies sample counts for evaluating the fingerprinting process ('1K samples drawn from pz for each fingerprint φ, and use 1K fingerprints'), but it does not provide explicit training, validation, or test dataset splits for these underlying image datasets. |
| Hardware Specification | Yes | For instance, the baseline method requires approximately one hour to fine-tune the model for each individual user. For a key capacity of 232 users, it would take an estimated 232 hours to train on an NVIDIA V100 GPU. Specifically, the average attribution time for the baseline method is approximately two seconds, while the proposed method takes approximately 126 seconds on average for 1k optimization trials (in parallel) using an NVIDIA V100 GPU. |
| Software Dependencies | No | The paper does not provide specific version numbers for software dependencies such as programming languages, libraries, or frameworks used in the experiments. |
| Experiment Setup | Yes | To compute the empirical accuracy (Eq. (1)), we use 1K samples drawn from pz for each fingerprint φ, and use 1K fingerprints where each bit is drawn independently from a Bernoulli distribution with p = 0.5. In Table 1, we show that both constraints on ˆα and parallel search with 20 initial guesses improve the empirical attribution accuracy across models and datasets. Noising adds a Gaussian white noise of standard deviation randomly sample from U[0, 0.1]. Blurring uses a randomly selected Gaussian kernel size from [3, 7, 9, 16, 25] and a standard deviation of [0.5, 1.0, 1.5, 2.0]. We randomly sample the JPEG quality from [80, 70, 60, 50]. |