The Uncanny Valley: A Comprehensive Analysis of Diffusion Models (2402.13369v1)
Abstract: Through Diffusion Models (DMs), we have made significant advances in generating high-quality images. Our exploration of these models delves deeply into their core operational principles by systematically investigating key aspects across various DM architectures: i) noise schedules, ii) samplers, and iii) guidance. Our comprehensive examination of these models sheds light on their hidden fundamental mechanisms, revealing the concealed foundational elements that are essential for their effectiveness. Our analyses emphasize the hidden key factors that determine model performance, offering insights that contribute to the advancement of DMs. Past findings show that the configuration of noise schedules, samplers, and guidance is vital to the quality of generated images; however, models reach a stable level of quality across different configurations at a remarkably similar point, revealing that the decisive factors for optimal performance predominantly reside in the diffusion process dynamics and the structural design of the model's network, rather than the specifics of configuration details. Our comparative analysis reveals that Denoising Diffusion Probabilistic Model (DDPM)-based diffusion dynamics consistently outperform the Noise Conditioned Score Network (NCSN)-based ones, not only when evaluated in their original forms but also when continuous through Stochastic Differential Equation (SDE)-based implementations.
- Numerical continuation methods: an introduction, volume 13. Springer Science & Business Media, 2012.
- Anderson, B. D. Reverse-time diffusion equation models. Stochastic Processes and their Applications, 12(3):313–326, 1982.
- Computer Methods for Ordinary Differential Equations and Differential-Algebraic Equations. Society for Industrial and Applied Mathematics, 1998.
- Analytic-dpm: an analytic estimate of the optimal reverse variance in diffusion probabilistic models. arXiv preprint arXiv:2201.06503, 2022.
- A note on the inception score. arXiv preprint arXiv:1801.01973, 2018.
- A study on the evaluation of generative models, 2022.
- Borji, A. Pros and cons of gan evaluation measures: New developments. Computer Vision and Image Understanding, 215:103329, 2022.
- On the design fundamentals of diffusion models: A survey. arXiv preprint arXiv:2306.04542, 2023.
- Chen, T. On the importance of noise scheduling for diffusion models. arXiv preprint arXiv:2301.10972, 2023.
- Analog bits: Generating discrete data using diffusion models with self-conditioning. arXiv preprint arXiv:2208.04202, 2022.
- Fast and accurate deep network learning by exponential linear units (elus). arXiv preprint arXiv:1511.07289, 2015.
- Diffusion models in vision: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.
- Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34:8780–8794, 2021.
- Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598, 2022.
- Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020.
- Estimation of non-normalized statistical models by score matching. Journal of Machine Learning Research, 6(4), 2005.
- Scalable adaptive computation for iterative generation. arXiv preprint arXiv:2212.11972, 2022.
- Feature likelihood score: Evaluating generalization of generative models using samples. arXiv preprint arXiv:2302.04440, 2023.
- Elucidating the design space of diffusion-based generative models. Advances in Neural Information Processing Systems, 35:26565–26577, 2022.
- Diffusion models in medical imaging: A comprehensive survey. Medical Image Analysis, pp. 102846, 2023.
- Variational diffusion models. Advances in neural information processing systems, 34:21696–21707, 2021.
- Numerical Solution of Stochastic Differential Equations, volume 23. Springer Science & Business Media, 2013.
- Improved precision and recall metric for assessing generative models. Advances in Neural Information Processing Systems, 32, 2019.
- Discrete predictor-corrector diffusion models for image synthesis. In The Eleventh International Conference on Learning Representations, 2022.
- Refinenet: Multi-path refinement networks for high-resolution semantic segmentation. pp. 1925–1934, 2017.
- Pseudo numerical methods for diffusion models on manifolds. arXiv preprint arXiv:2202.09778, 2022.
- Dpm-solver++: Fast solver for guided sampling of diffusion probabilistic models. arXiv preprint arXiv:2211.01095, 2022.
- Improved denoising diffusion probabilistic models. pp. 8162–8171, 2021.
- U-net: Convolutional networks for biomedical image segmentation. pp. 234–241, 2015.
- Pixelcnn++: Improving the pixelcnn with discretized logistic mixture likelihood and other modifications. arXiv preprint arXiv:1701.05517, 2017.
- Deep unsupervised learning using nonequilibrium thermodynamics. In International conference on machine learning, pp. 2256–2265. PMLR, 2015.
- Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502, 2020a.
- Generative modeling by estimating gradients of the data distribution. Advances in neural information processing systems, 32, 2019.
- Improved techniques for training score-based generative models. Advances in neural information processing systems, 33:12438–12448, 2020.
- Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020b.
- Attention is all you need. Advances in neural information processing systems, 30, 2017.
- Group normalization. In Proceedings of the European conference on computer vision (ECCV), pp. 3–19, 2018.
- Diffusion models: A comprehensive survey of methods and applications. ACM Computing Surveys, 56(4):1–39, 2023.
- Unipc: A unified predictor-corrector framework for fast sampling of diffusion models. arXiv preprint arXiv:2302.04867, 2023.
- Karam Ghanem (2 papers)
- Danilo Bzdok (17 papers)