AIConfPaper
← Search

Miika Aittala

16 accepted papers

2025

Elucidated Rolling Diffusion Models for Probabilistic Forecasting of Complex Dynamics

Salva Rühling Cachay, Miika Aittala, Karsten Kreis, Noah D Brenowitz, Arash Vahdat, Morteza Mardani +1
NeurIPS 2025poster

Diffusion models are a powerful tool for probabilistic forecasting, yet most applications in high-dimensional complex systems predict future states individually. This approach struggles to model complex temporal dependencies and fails to explicitly account for the progressive growth of uncertainty i…

Cited by 0Source ↗Scholar ↗
2024

Analyzing and Improving the Training Dynamics of Diffusion Models

Tero Karras, Miika Aittala, Jaakko Lehtinen, Janne Hellsten, Timo Aila, Samuli Laine
CVPR 2024poster

Diffusion models currently dominate the field of data-driven image synthesis with their unparalleled scaling to large datasets. In this paper we identify and rectify several causes for uneven and ineffective training in the popular ADM diffusion model architecture without altering its high-level str…

Cited by 146Source ↗PDF ↗code ↗Scholar ↗
2024

Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models

Tuomas Kynkäänniemi, Miika Aittala, Tero Karras, Samuli Laine, Timo Aila, Jaakko Lehtinen
NeurIPS 2024poster

Guidance is a crucial technique for extracting the best performance out of image-generating diffusion models. Traditionally, a constant guidance weight has been applied throughout the sampling chain of an image. We show that guidance is clearly harmful toward the beginning of the chain (high noise l…

Cited by 48Source ↗PDF ↗code ↗Scholar ↗
2024

Guiding a Diffusion Model with a Bad Version of Itself

Tero Karras, Miika Aittala, Tuomas Kynkäänniemi, Jaakko Lehtinen, Timo Aila, Samuli Laine
NeurIPS 2024oral

The primary axes of interest in image-generating diffusion models are image quality, the amount of variation in the results, and how well the results align with a given condition, e.g., a class label or a text prompt. The popular classifier-free guidance approach uses an unconditional model to guide…

Cited by 38Source ↗PDF ↗code ↗Scholar ↗
2023

Generative Novel View Synthesis with 3D-Aware Diffusion Models

Eric R. Chan, Koki Nagano, Matthew A. Chan, Alexander W. Bergman, Jeong Joon Park, Axel Levy +4
ICCV 2023oral

We present a diffusion-based model for 3D-aware generative novel view synthesis from as few as a single input image. Our model samples from the distribution of possible renderings consistent with the input and, even in the presence of ambiguity, is capable of rendering diverse and plausible novel vi…

Cited by 235PDF ↗code ↗Scholar ↗
2023

The Role of ImageNet Classes in Fréchet Inception Distance

Tuomas Kynkäänniemi, Tero Karras, Miika Aittala, Timo Aila, Jaakko Lehtinen
ICLR 2023top-25%

Fréchet Inception Distance (FID) is the primary metric for ranking models in data-driven generative modeling. While remarkably successful, the metric is known to sometimes disagree with human judgement. We investigate a root cause of these discrepancies, and visualize what FID "looks at" in generate…

Cited by 216Source ↗PDF ↗code ↗Scholar ↗
2022

Elucidating the Design Space of Diffusion-Based Generative Models

Tero Karras, Miika Aittala, Timo Aila, Samuli Laine
NeurIPS 2022accept

We argue that the theory and practice of diffusion-based generative models are currently unnecessarily convoluted and seek to remedy the situation by presenting a design space that clearly separates the concrete design choices. This lets us identify several changes to both the sampling and training…

Cited by 1989Source ↗PDF ↗code ↗Scholar ↗
2022

Generating Long Videos of Dynamic Scenes

Tim Brooks, Janne Hellsten, Miika Aittala, Ting-chun Wang, Timo Aila, Jaakko Lehtinen +3
NeurIPS 2022accept

We present a video generation model that accurately reproduces object motion, changes in camera viewpoint, and new content that arises over time. Existing video generation methods often fail to produce new content as a function of time while maintaining consistencies expected in real environments, s…

Cited by 121Source ↗PDF ↗Scholar ↗
2021

Alias-Free Generative Adversarial Networks

Tero Karras, Miika Aittala, Samuli Laine, Erik Härkönen, Janne Hellsten, Jaakko Lehtinen +1
NeurIPS 2021oral

We observe that despite their hierarchical convolutional nature, the synthesis process of typical generative adversarial networks depends on absolute pixel coordinates in an unhealthy manner. This manifests itself as, e.g., detail appearing to be glued to image coordinates instead of the surfaces of…

Cited by 1947Source ↗PDF ↗code ↗Scholar ↗
2021

What You Can Learn by Staring at a Blank Wall

Prafull Sharma, Miika Aittala, Yoav Y. Schechner, Antonio Torralba, Gregory W. Wornell, William T. Freeman +1
ICCV 2021poster

We present a passive non-line-of-sight method that infers the number of people or activity of a person from the observation of a blank wall in an unknown room. Our technique analyzes complex imperceptible changes in indirect illumination in a video of the wall to reveal a signal that is correlated w…

Cited by 19PDF ↗Scholar ↗
2020

Analyzing and Improving the Image Quality of StyleGAN

Tero Karras, Samuli Laine, Miika Aittala, Janne Hellsten, Jaakko Lehtinen, Timo Aila
CVPR 2020poster

The style-based GAN architecture (StyleGAN) yields state-of-the-art results in data-driven unconditional generative image modeling. We expose and analyze several of its characteristic artifacts, and propose changes in both model architecture and training methods to address them. In particular, we re…

Cited by 7789PDF ↗code ↗Scholar ↗
2020

Training Generative Adversarial Networks with Limited Data

Tero Karras, Miika Aittala, Janne Hellsten, Samuli Laine, Jaakko Lehtinen, Timo Aila
NeurIPS 2020oral

Training generative adversarial networks (GAN) using too little data typically leads to discriminator overfitting, causing training to diverge. We propose an adaptive discriminator augmentation mechanism that significantly stabilizes training in limited data regimes. The approach does not require ch…

Cited by 2393Source ↗PDF ↗code ↗Scholar ↗
2019

Computational Mirrors: Blind Inverse Light Transport by Deep Matrix Factorization

Miika Aittala, Prafull Sharma, Lukas Murmann, Adam Yedidia, Gregory Wornell, Bill Freeman +1
NeurIPS 2019poster

We recover a video of the motion taking place in a hidden scene by observing changes in indirect illumination in a nearby uncalibrated visible region. We solve this problem by factoring the observed video into a matrix product between the unknown hidden scene video and an unknown light transport mat…

Cited by 56Source ↗PDF ↗Scholar ↗
2018

Noise2Noise: Learning Image Restoration without Clean Data

Jaakko Lehtinen, Jacob Munkberg, Jon Hasselgren, Samuli Laine, Tero Karras, Miika Aittala +1
ICML 2018oral

We apply basic statistical reasoning to signal reconstruction by machine learning - learning to map corrupted observations to clean signals - with a simple and powerful conclusion: it is possible to learn to restore images by only looking at corrupted examples, at performance at and sometimes exceed…

Cited by 2089Source ↗PDF ↗code ↗Scholar ↗