Optical Diffusion Models for Image Generation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910679202529280 |
|---|---|
| author | Oguz, Ilker Dinc, Niyazi Ulas Yildirim, Mustafa Ke, Junjie Yoo, Innfarn Wang, Qifei Yang, Feng Moser, Christophe Psaltis, Demetri |
| author_facet | Oguz, Ilker Dinc, Niyazi Ulas Yildirim, Mustafa Ke, Junjie Yoo, Innfarn Wang, Qifei Yang, Feng Moser, Christophe Psaltis, Demetri |
| contents | Diffusion models generate new samples by progressively decreasing the noise from the initially provided random distribution. This inference procedure generally utilizes a trained neural network numerous times to obtain the final output, creating significant latency and energy consumption on digital electronic hardware such as GPUs. In this study, we demonstrate that the propagation of a light beam through a semi-transparent medium can be programmed to implement a denoising diffusion model on image samples. This framework projects noisy image patterns through passive diffractive optical layers, which collectively only transmit the predicted noise term in the image. The optical transparent layers, which are trained with an online training approach, backpropagating the error to the analytical model of the system, are passive and kept the same across different steps of denoising. Hence this method enables high-speed image generation with minimal power consumption, benefiting from the bandwidth and energy efficiency of optical information processing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_10897 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Optical Diffusion Models for Image Generation Oguz, Ilker Dinc, Niyazi Ulas Yildirim, Mustafa Ke, Junjie Yoo, Innfarn Wang, Qifei Yang, Feng Moser, Christophe Psaltis, Demetri Optics Computer Vision and Pattern Recognition Machine Learning Diffusion models generate new samples by progressively decreasing the noise from the initially provided random distribution. This inference procedure generally utilizes a trained neural network numerous times to obtain the final output, creating significant latency and energy consumption on digital electronic hardware such as GPUs. In this study, we demonstrate that the propagation of a light beam through a semi-transparent medium can be programmed to implement a denoising diffusion model on image samples. This framework projects noisy image patterns through passive diffractive optical layers, which collectively only transmit the predicted noise term in the image. The optical transparent layers, which are trained with an online training approach, backpropagating the error to the analytical model of the system, are passive and kept the same across different steps of denoising. Hence this method enables high-speed image generation with minimal power consumption, benefiting from the bandwidth and energy efficiency of optical information processing. |
| title | Optical Diffusion Models for Image Generation |
| topic | Optics Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2407.10897 |