Scaling Rectified Flow Transformers for High-Resolution Image Synthesis
- URL: http://arxiv.org/abs/2403.03206v1
- Date: Tue, 5 Mar 2024 18:45:39 GMT
- Title: Scaling Rectified Flow Transformers for High-Resolution Image Synthesis
- Authors: Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas
M\"uller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel,
Dustin Podell, Tim Dockhorn, Zion English, Kyle Lacey, Alex Goodwin, Yannik
Marek, Robin Rombach
- Abstract summary: Rectified flow is a recent generative model formulation that connects data and noise in a straight line.
We improve existing noise sampling techniques for training rectified flow models by biasing them towards perceptually relevant scales.
We present a novel transformer-based architecture for text-to-image generation that uses separate weights for the two modalities.
- Score: 22.11487736315616
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Diffusion models create data from noise by inverting the forward paths of
data towards noise and have emerged as a powerful generative modeling technique
for high-dimensional, perceptual data such as images and videos. Rectified flow
is a recent generative model formulation that connects data and noise in a
straight line. Despite its better theoretical properties and conceptual
simplicity, it is not yet decisively established as standard practice. In this
work, we improve existing noise sampling techniques for training rectified flow
models by biasing them towards perceptually relevant scales. Through a
large-scale study, we demonstrate the superior performance of this approach
compared to established diffusion formulations for high-resolution
text-to-image synthesis. Additionally, we present a novel transformer-based
architecture for text-to-image generation that uses separate weights for the
two modalities and enables a bidirectional flow of information between image
and text tokens, improving text comprehension, typography, and human preference
ratings. We demonstrate that this architecture follows predictable scaling
trends and correlates lower validation loss to improved text-to-image synthesis
as measured by various metrics and human evaluations. Our largest models
outperform state-of-the-art models, and we will make our experimental data,
code, and model weights publicly available.
Related papers
- Is Synthetic Image Useful for Transfer Learning? An Investigation into Data Generation, Volume, and Utilization [62.157627519792946]
We introduce a novel framework called bridged transfer, which initially employs synthetic images for fine-tuning a pre-trained model to improve its transferability.
We propose dataset style inversion strategy to improve the stylistic alignment between synthetic and real images.
Our proposed methods are evaluated across 10 different datasets and 5 distinct models, demonstrating consistent improvements.
arXiv Detail & Related papers (2024-03-28T22:25:05Z) - ReNoise: Real Image Inversion Through Iterative Noising [62.96073631599749]
We introduce an inversion method with a high quality-to-operation ratio, enhancing reconstruction accuracy without increasing the number of operations.
We evaluate the performance of our ReNoise technique using various sampling algorithms and models, including recent accelerated diffusion models.
arXiv Detail & Related papers (2024-03-21T17:52:08Z) - FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis [48.9652334528436]
We introduce an innovative, training-free approach FouriScale from the perspective of frequency domain analysis.
We replace the original convolutional layers in pre-trained diffusion models by incorporating a dilation technique along with a low-pass operation.
Our method successfully balances the structural integrity and fidelity of generated images, achieving an astonishing capacity of arbitrary-size, high-resolution, and high-quality generation.
arXiv Detail & Related papers (2024-03-19T17:59:33Z) - Denoising Autoregressive Representation Learning [13.185567468951628]
Our method, DARL, employs a decoder-only Transformer to predict image patches autoregressively.
We show that the learned representation can be improved by using tailored noise schedules and longer training in larger models.
arXiv Detail & Related papers (2024-03-08T10:19:00Z) - Steered Diffusion: A Generalized Framework for Plug-and-Play Conditional
Image Synthesis [62.07413805483241]
Steered Diffusion is a framework for zero-shot conditional image generation using a diffusion model trained for unconditional generation.
We present experiments using steered diffusion on several tasks including inpainting, colorization, text-guided semantic editing, and image super-resolution.
arXiv Detail & Related papers (2023-09-30T02:03:22Z) - Stage-by-stage Wavelet Optimization Refinement Diffusion Model for
Sparse-View CT Reconstruction [14.037398189132468]
We present an innovative approach named the Stage-by-stage Wavelet Optimization Refinement Diffusion (SWORD) model for sparse-view CT reconstruction.
Specifically, we establish a unified mathematical model integrating low-frequency and high-frequency generative models, achieving the solution with optimization procedure.
Our method rooted in established optimization theory, comprising three distinct stages, including low-frequency generation, high-frequency refinement and domain transform.
arXiv Detail & Related papers (2023-08-30T10:48:53Z) - Training on Thin Air: Improve Image Classification with Generated Data [28.96941414724037]
Diffusion Inversion is a simple yet effective method to generate diverse, high-quality training data for image classification.
Our approach captures the original data distribution and ensures data coverage by inverting images to the latent space of Stable Diffusion.
We identify three key components that allow our generated images to successfully supplant the original dataset.
arXiv Detail & Related papers (2023-05-24T16:33:02Z) - IRGen: Generative Modeling for Image Retrieval [82.62022344988993]
In this paper, we present a novel methodology, reframing image retrieval as a variant of generative modeling.
We develop our model, dubbed IRGen, to address the technical challenge of converting an image into a concise sequence of semantic units.
Our model achieves state-of-the-art performance on three widely-used image retrieval benchmarks and two million-scale datasets.
arXiv Detail & Related papers (2023-03-17T17:07:36Z) - Person Image Synthesis via Denoising Diffusion Model [116.34633988927429]
We show how denoising diffusion models can be applied for high-fidelity person image synthesis.
Our results on two large-scale benchmarks and a user study demonstrate the photorealism of our proposed approach under challenging scenarios.
arXiv Detail & Related papers (2022-11-22T18:59:50Z) - High-Fidelity Synthesis with Disentangled Representation [60.19657080953252]
We propose an Information-Distillation Generative Adrial Network (ID-GAN) for disentanglement learning and high-fidelity synthesis.
Our method learns disentangled representation using VAE-based models, and distills the learned representation with an additional nuisance variable to the separate GAN-based generator for high-fidelity synthesis.
Despite the simplicity, we show that the proposed method is highly effective, achieving comparable image generation quality to the state-of-the-art methods using the disentangled representation.
arXiv Detail & Related papers (2020-01-13T14:39:40Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.