1.5k citations · 8.4k across the 21 of their papers we have counts for
24 papers
Finite Difference Flow Optimization for RL Post-Training of Text-to-Image Models
David McAllister, Miika Aittala, Tero Karras +4
Reinforcement learning (RL) has become a standard technique for post-training diffusion-based image synthesis models, as it enables learning from reward signals to explicitly impro…
Guiding a Diffusion Model with a Bad Version of Itself
Tero Karras, Miika Aittala, Tuomas Kynkäänniemi +3
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,…
Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models
Tuomas Kynkäänniemi, Miika Aittala, Tero Karras +3
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…
Analyzing and Improving the Training Dynamics of Diffusion Models
Tero Karras, Miika Aittala, Jaakko Lehtinen +3
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 cause…
StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesis
Axel Sauer, Tero Karras, Samuli Laine +2
Text-to-image synthesis has recently seen significant progress thanks to large pretrained language models, large-scale training data, and the introduction of scalable model familie…
eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers
Yogesh Balaji, Seungjun Nah, Xun Huang +10
Large-scale diffusion-based generative models have led to breakthroughs in text-conditioned high-resolution image synthesis. Starting from random noise, such text-to-image diffusio…
Disentangling Random and Cyclic Effects in Time-Lapse Sequences
Erik Härkönen, Miika Aittala, Tuomas Kynkäänniemi +3
Time-lapse image sequences offer visually compelling insights into dynamic processes that are too slow to observe in real time. However, playing a long time-lapse sequence back as…
Generating Long Videos of Dynamic Scenes
Tim Brooks, Janne Hellsten, Miika Aittala +6
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…
Elucidating the Design Space of Diffusion-Based Generative Models
Tero Karras, Miika Aittala, Timo Aila +1
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…
The Role of ImageNet Classes in Fréchet Inception Distance
Tuomas Kynkäänniemi, Tero Karras, Miika Aittala +2
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…
Alias-Free Generative Adversarial Networks
Tero Karras, Miika Aittala, Samuli Laine +4
We observe that despite their hierarchical convolutional nature, the synthesis process of typical generative adversarial networks depends on absolute pixel coordinates in an unheal…
Modular Primitives for High-Performance Differentiable Rendering
Samuli Laine, Janne Hellsten, Tero Karras +3
We present a modular differentiable renderer design that yields performance superior to previous methods by leveraging existing, highly optimized hardware graphics pipelines. Our d…
Training Generative Adversarial Networks with Limited Data
Tero Karras, Miika Aittala, Janne Hellsten +3
Training generative adversarial networks (GAN) using too little data typically leads to discriminator overfitting, causing training to diverge. We propose an adaptive discriminator…
Analyzing and Improving the Image Quality of StyleGAN
Tero Karras, Samuli Laine, Miika Aittala +3
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 characteris…
Semi-supervised semantic segmentation needs strong, varied perturbations
Geoff French, Samuli Laine, Timo Aila +2
Consistency regularization describes a class of approaches that have yielded ground breaking results in semi-supervised classification problems. Prior work has established the clus…
Few-Shot Unsupervised Image-to-Image Translation
Ming-Yu Liu, Xun Huang, Arun Mallya +4
Unsupervised image-to-image translation methods learn to map images in a given class to an analogous image in a different class, drawing on unstructured (non-registered) datasets o…
Improved Precision and Recall Metric for Assessing Generative Models
Tuomas Kynkäänniemi, Tero Karras, Samuli Laine +2
The ability to automatically estimate the quality and coverage of the samples produced by a generative model is a vital requirement for driving algorithm research. We present an ev…
High-Quality Self-Supervised Deep Image Denoising
Samuli Laine, Tero Karras, Jaakko Lehtinen +1
We describe a novel method for training high-quality image denoising models based on unorganized collections of corrupted images. The training does not need access to clean referen…
A Style-Based Generator Architecture for Generative Adversarial Networks
Tero Karras, Samuli Laine, Timo Aila
We propose an alternative generator architecture for generative adversarial networks, borrowing from style transfer literature. The new architecture leads to an automatically learn…
Noise2Noise: Learning Image Restoration without Clean Data
Jaakko Lehtinen, Jacob Munkberg, Jon Hasselgren +4
We apply basic statistical reasoning to signal reconstruction by machine learning -- learning to map corrupted observations to clean signals -- with a simple and powerful conclusio…
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Tero Karras, Timo Aila, Samuli Laine +1
We describe a new training methodology for generative adversarial networks. The key idea is to grow both the generator and discriminator progressively: starting from a low resoluti…
Pruning Convolutional Neural Networks for Resource Efficient Inference
Pavlo Molchanov, Stephen Tyree, Tero Karras +2
We propose a new formulation for pruning convolutional kernels in neural networks to enable efficient inference. We interleave greedy criteria-based pruning with fine-tuning by bac…
Temporal Ensembling for Semi-Supervised Learning
Samuli Laine, Timo Aila
In this paper, we present a simple and efficient method for training deep neural networks in a semi-supervised setting where only a small portion of training data is labeled. We in…
Production-Level Facial Performance Capture Using Deep Convolutional Neural Networks
Samuli Laine, Tero Karras, Timo Aila +5
We present a real-time deep learning framework for video-based facial performance capture -- the dense 3D tracking of an actor's face given a monocular video. Our pipeline begins w…