Download pdf
Advancements in Diffusion Models for Image Generation: A Comparative Analysis of DDPM, LDM, and DDIM
- 1 School of information, Xiamen University, Xiamen, China
* Author to whom correspondence should be addressed.
Abstract
This research provides a thorough exploration of diffusion models in image generation, comparing various methodologies to assess their efficacy and efficiency. The study begins with an introduction to foundational technologies and key concepts, progressing through an analysis of basic and advanced models, including Latent Diffusion Models (LDMs), Denoising Diffusion Implicit Models (DDIMs), and control models. The research evaluates these models based on their performance, computational efficiency, and future development potential. The review details the evolution of diffusion models from early stochastic processes to their current status as advanced generative models. Key principles, such as iterative noise addition and removal, are examined to understand the transformation from simple distributions to complex data representations. Innovations enhancing model efficiency, including advancements in score matching and neural network integration, are discussed. A thorough comparative analysis highlights the strengths and limitations of each model. The study identifies ongoing challenges such as interpretability and computational cost and proposes future research directions to address these issues. The findings aim to guide researchers and practitioners in advancing diffusion model technologies, offering insights into their impact on image generation and potential future developments.
Keywords
Diffusion Models, Image Generation, Denoising Diffusion Implicit Models.
[1]. Ho J Jain A Abbeel P 2020 Denoising diffusion probabilistic models Advances in neural information processing systems vol 33 pp 6840-6851
[2]. Jascha S D Weiss E Maheswaranathan N and Ganguli S 2015 Deep unsupervised learning using nonequilibrium thermodynamics In International Conference on Machine Learning pp 2256–2265
[3]. De Groot S R Mazur P 2013 Non-equilibrium thermodynamics Courier Corporation
[4]. Song Y Ermon S 2019 Generative modeling by estimating gradients of the data distribution Advances in neural information processing systems vol 32
[5]. Austin J Johnson D D Ho J et al. 2021 Structured denoising diffusion models in discrete state-spaces Advances in Neural Information Processing Systems vol 34 pp 17981-17993
[6]. Weiss T Mayo Yanes E Chakraborty S et al. 2023 Guided diffusion for inverse molecular design Nature Computational Science vol 3 no 10 pp 873-882
[7]. Cao C Cui Z X Wang Y et al. 2024 High-frequency space diffusion model for accelerated mri IEEE Transactions on Medical Imaging
[8]. Kingma D et al. 2021 Variational diffusion models Advances in neural information processing systems vol 34 pp 21696-21707
[9]. Dhariwal P and Alexander N 2021 Diffusion models beat gans on image synthesis Advances in neural information processing systems vol 34 pp 8780-8794
[10]. Zhang L Anyi R and Maneesh A 2023 Adding conditional control to text-to-image diffusion models Proceedings of the IEEE/CVF International Conference on Computer Vision pp 3836-3847
[11]. Rombach R et al. 2022 High-resolution image synthesis with latent diffusion models Proceedings of the IEEE/CVF conference on computer vision and pattern recognition pp 10684-10695
[12]. Song J Meng C Ermon S 2020 Denoising diffusion implicit models arxiv preprint 2010.02502
Cite this article
Jin,Z. (2024). Advancements in Diffusion Models for Image Generation: A Comparative Analysis of DDPM, LDM, and DDIM. Applied and Computational Engineering,104,96-103.
Data availability
The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.
Disclaimer/Publisher's Note
The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of EWA Publishing and/or the editor(s). EWA Publishing and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
About volume
Volume title: Proceedings of the 2nd International Conference on Machine Learning and Automation
© 2024 by the author(s). Licensee EWA Publishing, Oxford, UK. This article is an open access article distributed under the terms and
conditions of the Creative Commons Attribution (CC BY) 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 policy for details).