How Many Images Does It Take? Estimating Imitation Thresholds in Text-to-Image Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909981819797504 |
|---|---|
| author | Verma, Sahil Rassin, Royi Das, Arnav Bhatt, Gantavya Seshadri, Preethi Shah, Chirag Bilmes, Jeff Hajishirzi, Hannaneh Elazar, Yanai |
| author_facet | Verma, Sahil Rassin, Royi Das, Arnav Bhatt, Gantavya Seshadri, Preethi Shah, Chirag Bilmes, Jeff Hajishirzi, Hannaneh Elazar, Yanai |
| contents | Text-to-image models are trained using large datasets of image-text pairs collected from the internet. These datasets often include copyrighted and private images. Training models on such datasets enables them to generate images that might violate copyright laws and individual privacy. This phenomenon is termed imitation -- generation of images with content that has recognizable similarity to its training images. In this work we estimate the point at which a model was trained on enough instances of a concept to be able to imitate it -- the imitation threshold. We posit this question as a new problem and propose an efficient approach that estimates the imitation threshold without incurring the colossal cost of training these models from scratch. We experiment with two domains -- human faces and art styles, and evaluate four text-to-image models that were trained on three pretraining datasets. We estimate the imitation threshold of these models to be in the range of 200-700 images, depending on the domain and the model. The imitation threshold provides an empirical basis for copyright violation claims and acts as a guiding principle for text-to-image model developers that aim to comply with copyright and privacy laws. Website: https://how-many-van-goghs-does-it-take.github.io/. Code: https://github.com/vsahil/MIMETIC-2. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_15002 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | How Many Images Does It Take? Estimating Imitation Thresholds in Text-to-Image Models Verma, Sahil Rassin, Royi Das, Arnav Bhatt, Gantavya Seshadri, Preethi Shah, Chirag Bilmes, Jeff Hajishirzi, Hannaneh Elazar, Yanai Computer Vision and Pattern Recognition Text-to-image models are trained using large datasets of image-text pairs collected from the internet. These datasets often include copyrighted and private images. Training models on such datasets enables them to generate images that might violate copyright laws and individual privacy. This phenomenon is termed imitation -- generation of images with content that has recognizable similarity to its training images. In this work we estimate the point at which a model was trained on enough instances of a concept to be able to imitate it -- the imitation threshold. We posit this question as a new problem and propose an efficient approach that estimates the imitation threshold without incurring the colossal cost of training these models from scratch. We experiment with two domains -- human faces and art styles, and evaluate four text-to-image models that were trained on three pretraining datasets. We estimate the imitation threshold of these models to be in the range of 200-700 images, depending on the domain and the model. The imitation threshold provides an empirical basis for copyright violation claims and acts as a guiding principle for text-to-image model developers that aim to comply with copyright and privacy laws. Website: https://how-many-van-goghs-does-it-take.github.io/. Code: https://github.com/vsahil/MIMETIC-2. |
| title | How Many Images Does It Take? Estimating Imitation Thresholds in Text-to-Image Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.15002 |