Straighten Viscous Rectified Flow via Noise Optimization

• URL: http://arxiv.org/abs/2507.10218v1
• Date: Mon, 14 Jul 2025 12:35:17 GMT
• Title: Straighten Viscous Rectified Flow via Noise Optimization
• Authors: Jimin Dai, Jiexi Yan, Jian Yang, Lei Luo,
• Abstract summary: The Reflow operation aims to straighten the inference trajectories of the rectified flow during training by constructing deterministic couplings between noises and images.<n>We identify critical limitations in Reflow, particularly its inability to rapidly generate high-quality images due to a distribution gap between images in its constructed deterministic couplings and real images.<n>We propose a novel alternative called Straighten Viscous Rectified Flow via Noise Optimization (VRFNO), which is a joint training framework integrating an encoder and a neural velocity field.
• Score: 24.065483360595458
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The Reflow operation aims to straighten the inference trajectories of the rectified flow during training by constructing deterministic couplings between noises and images, thereby improving the quality of generated images in single-step or few-step generation. However, we identify critical limitations in Reflow, particularly its inability to rapidly generate high-quality images due to a distribution gap between images in its constructed deterministic couplings and real images. To address these shortcomings, we propose a novel alternative called Straighten Viscous Rectified Flow via Noise Optimization (VRFNO), which is a joint training framework integrating an encoder and a neural velocity field. VRFNO introduces two key innovations: (1) a historical velocity term that enhances trajectory distinction, enabling the model to more accurately predict the velocity of the current trajectory, and (2) the noise optimization through reparameterization to form optimized couplings with real images which are then utilized for training, effectively mitigating errors caused by Reflow's limitations. Comprehensive experiments on synthetic data and real datasets with varying resolutions show that VRFNO significantly mitigates the limitations of Reflow, achieving state-of-the-art performance in both one-step and few-step generation tasks.

Related papers

• Free Lunch for Stabilizing Rectified Flow Inversion [11.80912018629953]
  Rectified-Flow (RF)-based generative models have emerged as strong alternatives to traditional diffusion models.<n>We propose Proximal-Mean Inversion (PMI), a training-free gradient correction method.<n>We also introduce mimic-CFG, a lightweight velocity correction scheme for editing tasks.
  arXiv  Detail & Related papers  (2026-02-12T11:42:36Z)
• FlowConsist: Make Your Flow Consistent with Real Trajectory [99.22869983378062]
  We argue that current fast-flow training paradigms suffer from two fundamental issues.<n> conditional velocities constructed from randomly paired noise-data samples introduce systematic trajectory drift.<n>We propose FlowConsist, a training framework designed to enforce trajectory consistency in fast flows.
  arXiv  Detail & Related papers  (2026-02-06T03:24:23Z)
• RenderFlow: Single-Step Neural Rendering via Flow Matching [17.56739408578129]
  We present a novel end-to-end, deterministic, single-step neural rendering framework, RenderFlow, built upon a flow matching paradigm.<n>Our method significantly accelerates rendering process and enhances both the physical plausibility and overall visual quality of the output.<n>The resulting pipeline achieves near real-time performance with photorealistic rendering quality, effectively bridging the gap between the efficiency of modern generative models and the precision of traditional physically based rendering.
  arXiv  Detail & Related papers  (2026-01-11T14:28:46Z)
• Balanced conic rectified flow [19.226787997122987]
  Rectified flow is a generative model that learns smooth transport mappings between two distributions through an ordinary differential equation (ODE)<n>In this work, we experimentally expose the limitations of the original rectified flow and propose a novel approach that incorporates real images into the training process.
  arXiv  Detail & Related papers  (2025-10-29T07:06:01Z)
• BADiff: Bandwidth Adaptive Diffusion Model [55.10134744772338]
  Traditional diffusion models produce high-fidelity images by performing a fixed number of denoising steps, regardless of downstream transmission limitations.<n>In practical cloud-to-device scenarios, limited bandwidth often necessitates heavy compression, leading to loss of fine textures and wasted computation.<n>We introduce a joint end-to-end training strategy where the diffusion model is conditioned on a target quality level derived from the available bandwidth.
  arXiv  Detail & Related papers  (2025-10-24T11:50:03Z)
• Exploring Fourier Prior and Event Collaboration for Low-Light Image Enhancement [1.8724535169356553]
  Event cameras provide performance gain for low-light image enhancement.<n>Currently, existing event-based methods feed a frame and events directly into a single model.<n>We propose a visibility restoration network with amplitude-phase entanglement.<n>In the second stage, a fusion strategy with dynamic alignment is proposed to mitigate the spatial mismatch.
  arXiv  Detail & Related papers  (2025-08-01T04:25:00Z)
• Solving Inverse Problems with FLAIR [59.02385492199431]
  Flow-based latent generative models are able to generate images with remarkable quality, even enabling text-to-image generation.<n>We present FLAIR, a novel training free variational framework that leverages flow-based generative models as a prior for inverse problems.<n>Results on standard imaging benchmarks demonstrate that FLAIR consistently outperforms existing diffusion- and flow-based methods in terms of reconstruction quality and sample diversity.
  arXiv  Detail & Related papers  (2025-06-03T09:29:47Z)
• NAMI: Efficient Image Generation via Progressive Rectified Flow Transformers [11.087309945227826]
  Flow-based transformer models for image generation have achieved state-of-the-art performance with larger model parameters, but their inference deployment cost remains high.<n>To enhance inference performance while maintaining generation quality, we propose progressive rectified flow transformers.<n>Experiments demonstrate that our approach achieves fast convergence and reduces inference time while ensuring generation quality.
  arXiv  Detail & Related papers  (2025-03-12T10:38:58Z)
• One-Step Diffusion Model for Image Motion-Deblurring [85.76149042561507]
  We propose a one-step diffusion model for deblurring (OSDD), a novel framework that reduces the denoising process to a single step.<n>To tackle fidelity loss in diffusion models, we introduce an enhanced variational autoencoder (eVAE), which improves structural restoration.<n>Our method achieves strong performance on both full and no-reference metrics.
  arXiv  Detail & Related papers  (2025-03-09T09:39:57Z)
• Learning Efficient and Effective Trajectories for Differential Equation-based Image Restoration [59.744840744491945]
  In this paper, we reformulate the trajectory optimization of this kind of method, focusing on enhancing both reconstruction quality and efficiency.<n>To mitigate the considerable computational burden associated with iterative sampling, we propose cost-aware trajectory distillation.<n>We fine-tune a foundational diffusion model (FLUX) with 12B parameters by using our algorithms, producing a unified framework for handling 7 kinds of image restoration tasks.
  arXiv  Detail & Related papers  (2024-10-07T07:46:08Z)
• Timestep-Aware Diffusion Model for Extreme Image Rescaling [47.89362819768323]
  We propose a novel framework called Timestep-Aware Diffusion Model (TADM) for extreme image rescaling.<n>TADM performs rescaling operations in the latent space of a pre-trained autoencoder.<n>It effectively leverages powerful natural image priors learned by a pre-trained text-to-image diffusion model.
  arXiv  Detail & Related papers  (2024-08-17T09:51:42Z)
• Motion-prior Contrast Maximization for Dense Continuous-Time Motion Estimation [34.529280562470746]
  We introduce a novel self-supervised loss combining the Contrast Maximization framework with a non-linear motion prior in the form of pixel-level trajectories. Their effectiveness is demonstrated in two scenarios: In dense continuous-time motion estimation, our method improves the zero-shot performance of a synthetically trained model by 29%.
  arXiv  Detail & Related papers  (2024-07-15T15:18:28Z)
• FlowIE: Efficient Image Enhancement via Rectified Flow [71.6345505427213]
  FlowIE is a flow-based framework that estimates straight-line paths from an elementary distribution to high-quality images. Our contributions are rigorously validated through comprehensive experiments on synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2024-06-01T17:29:29Z)
• SpikeReveal: Unlocking Temporal Sequences from Real Blurry Inputs with Spike Streams [44.02794438687478]
  Spike cameras have proven effective in capturing motion features and beneficial for solving this ill-posed problem. Existing methods fall into the supervised learning paradigm, which suffers from notable performance degradation when applied to real-world scenarios. We propose the first self-supervised framework for the task of spike-guided motion deblurring.
  arXiv  Detail & Related papers  (2024-03-14T15:29:09Z)
• Efficient Diffusion Model for Image Restoration by Residual Shifting [63.02725947015132]
  This study proposes a novel and efficient diffusion model for image restoration. Our method avoids the need for post-acceleration during inference, thereby avoiding the associated performance deterioration. Our method achieves superior or comparable performance to current state-of-the-art methods on three classical IR tasks.
  arXiv  Detail & Related papers  (2024-03-12T05:06:07Z)
• DGNet: Dynamic Gradient-Guided Network for Water-Related Optics Image Enhancement [77.0360085530701]
  Underwater image enhancement (UIE) is a challenging task due to the complex degradation caused by underwater environments. Previous methods often idealize the degradation process, and neglect the impact of medium noise and object motion on the distribution of image features. Our approach utilizes predicted images to dynamically update pseudo-labels, adding a dynamic gradient to optimize the network's gradient space.
  arXiv  Detail & Related papers  (2023-12-12T06:07:21Z)
• Single Image Optical Flow Estimation with an Event Camera [38.92408855196647]
  Event cameras are bio-inspired sensors that report intensity changes in microsecond resolution. We propose a single image (potentially blurred) and events based optical flow estimation approach.
  arXiv  Detail & Related papers  (2020-04-01T11:28:30Z)

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.