Theoretical and Natural Science
- The Open Access Proceedings Series for Conferences
Theoretical and Natural Science
Vol. 18, 08 December 2023
Open
Access | Article
Denoising diffusion model as handwritten digit generator
* Author to whom correspondence should be addressed.
Abstract
The diffusion model is the process by which variables are introduced and propagated through a population. The equation can be applied to any physical system that exhibits such processes through time. Generally speaking, diffusion occurs when a physical system is connected to another by a small number of interconnected points; the interconnected points' overall connectivity can be represented by a network. The diffusion model is a mathematically defined process used to analyse the movement of particles from a region of high concentration to that of low concentration. Based on the diffusion coefficient and polarization, several studies have established that the equilibrium port-to-port distance can be calculated. The diffusion model is useful for solving the problem of noise in imaging systems, especially when an object has similar properties in all directions. When discussing diffusion, it is essential to refer to the diffusion coefficient. The literatures find denoising diffusion model to involves the process where a pixel value is estimated based on values at surrounding pixels. On the other hand, a forward process is passing through an image and replacing pixels based on their quality estimates. Reconstruction involves reconstructing an image from its components, including the subsamples and low-quality components. This model achieves satisfactory performance on digital number image generation.
Keywords
deep learning, diffusion model, image generation, image reconstruction
References
1. erma-Usabiaga, G., Mukherjee, P., Perry, M. L., & Wandell, B. A. (2020). Data-science ready, multisite, human diffusion MRI white-matter-tract statistics. Scientific data, 7(1), 1-9.
2. Ho, J., Jain, A., & Abbeel, P. (2020). Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems, 33, 6840-6851.
3. Lin, H., Pennycook, G., & Rand, D. (2022). Thinking more or thinking differently? Using drift-diffusion modeling to illuminate why accuracy prompts decrease misinformation sharing.
4. Harvey, W., Naderiparizi, S., Masrani, V., Weilbach, C., & Wood, F. (2022). Flexible Diffusion Modeling of Long Videos. arXiv preprint arXiv:2205.11495.
5. Peng, C., Guo, P., Zhou, S. K., Patel, V. M., & Chellappa, R. (2022). Towards performant and reliable undersampled MR reconstruction via diffusion model sampling. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 623-633.
6. Pourhakkak, P., Taghizadeh, A., Taghizadeh, M., Ghaedi, M., & Haghdoust, S. (2021). Fundamentals of adsorption technology. In Interface Science and Technology 33, 1-70.
7. Kim, D., Shin, S., Song, K., Kang, W., & Moon, I. C. (2022). Soft truncation: A universal training technique of score-based diffusion model for high precision score estimation. In International Conference on Machine Learning, 11201-11228.
8. Liu, X., Yeo, K., & Lu, S. (2022). Statistical modeling for spatio-temporal data from stochastic convection-diffusion processes. Journal of the American Statistical Association, 117(539), 1482-1499
9. Song, Y., Durkan, C., Murray, I., & Ermon, S. (2021). Maximum likelihood training of score-based diffusion models. Advances in Neural Information Processing Systems, 34, 1415-1428
10. Capuani, S., & Palombo, M. (2020). Mini review on anomalous diffusion by MRI: potential advantages, pitfalls, limitations, nomenclature, and correct interpretation of literature. Frontiers in Physics, 7, 248
11. Firdaniza, F., Ruchjana, B. N., Chaerani, D., & Radianti, J. (2021). Information Diffusion Model in Twitter: A Systematic Literature Review. Information, 13(1), 13.
12. Kawar, B., Elad, M., Ermon, S., & Song, J. (2022). Denoising diffusion restoration models. arXiv preprint arXiv:2201.11793.
13. Kong, X., Gu, Z., & Yin, L. (2020, July). A unified information diffusion model for social networks. In 2020 IEEE Fifth International Conference on Data Science in Cyberspace (DSC), 38-44
14. Akhremenko, A. S., Stukal, D. K., & Petrov, A. P. (2020). Network vs message in protest diffusion on social media: theoretical and data analytics perspectives. Polis. Political Studies, 2(2), 73-91.
15. Choi, J., Lee, J., Shin, C., Kim, S., Kim, H., & Yoon, S. (2022). Perception Prioritized Training of Diffusion Models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 11472-11481.
16. Dhariwal, P., & Nichol, A. (2021). Diffusion models beat gans on image synthesis. Advances in Neural Information Processing Systems, 34, 8780-8794.
Data Availability
The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. Authors who publish this series agree to the following terms:
1. Authors retain copyright and grant the series right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this series.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the series's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this series.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See Open Access Instruction).
- Volume Title
- Proceedings of the 2nd International Conference on Computing Innovation and Applied Physics
- ISBN (Print)
- 978-1-83558-201-5
- ISBN (Online)
- 978-1-83558-202-2
- Published Date
- 08 December 2023
- Series
- Theoretical and Natural Science
- ISSN (Print)
- 2753-8818
- ISSN (Online)
- 2753-8826
- DOI
- 10.54254/2753-8818/18/20230353
- Copyright
- 08 December 2023
- Open Access
- This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited