Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SANA I2I: A Text Free Flow Matching Framework for Paired Image to Image Translation with a Case Study in Fetal MRI Artifact Reduction

Italo Felix Santos, Gilson Antonio Giraldi, Heron Werner Junior

Abstract

We propose SANA-I2I, a text-free high-resolution image-to-image generation framework that extends the SANA family by removing textual conditioning entirely. In contrast to SanaControlNet, which combines text and image-based control, SANA-I2I relies exclusively on paired source-target images to learn a conditional flow-matching model in latent space. The model learns a conditional velocity field that maps a target image distribution to another one, enabling supervised image translation without reliance on language prompts. We evaluate the proposed approach on the challenging task of fetal MRI motion artifact reduction. To enable paired training in this application, where real paired data are difficult to acquire, we adopt a synthetic data generation strategy based on the method proposed by Duffy et al., which simulates realistic motion artifacts in fetal magnetic resonance imaging (MRI). Experimental results demonstrate that SANA-I2I effectively suppresses motion artifacts while preserving anatomical structure, achieving competitive performance few inference steps. These results highlight the efficiency and suitability of our proposed flow-based, text-free generative models for supervised image-to-image tasks in medical imaging.

SANA I2I: A Text Free Flow Matching Framework for Paired Image to Image Translation with a Case Study in Fetal MRI Artifact Reduction

Abstract

We propose SANA-I2I, a text-free high-resolution image-to-image generation framework that extends the SANA family by removing textual conditioning entirely. In contrast to SanaControlNet, which combines text and image-based control, SANA-I2I relies exclusively on paired source-target images to learn a conditional flow-matching model in latent space. The model learns a conditional velocity field that maps a target image distribution to another one, enabling supervised image translation without reliance on language prompts. We evaluate the proposed approach on the challenging task of fetal MRI motion artifact reduction. To enable paired training in this application, where real paired data are difficult to acquire, we adopt a synthetic data generation strategy based on the method proposed by Duffy et al., which simulates realistic motion artifacts in fetal magnetic resonance imaging (MRI). Experimental results demonstrate that SANA-I2I effectively suppresses motion artifacts while preserving anatomical structure, achieving competitive performance few inference steps. These results highlight the efficiency and suitability of our proposed flow-based, text-free generative models for supervised image-to-image tasks in medical imaging.

Paper Structure

This paper contains 15 sections, 2 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Diffusion and Flow-Based Generative Models
  4. Image-Conditioned Generation and Control Mechanisms
  5. Motion Artifact Simulation and Fetal MRI Correction
  6. Methodology
  7. From SanaControlNet to Text-Free SANA-I2I
  8. Data Description
  9. Training
  10. Results
  11. Quantitative Evaluation on Synthetic Data
  12. Evaluation on Real Corrupted Images
  13. Qualitative Analysis
  14. Ablation Study
  15. Final Remarks

Figures (5)

  • Figure 1: Comparison between the SanaControlNet pipeline and the proposed text-free SANA-I2I architecture.
  • Figure 2: Qualitative comparison between the proposed method and CycleGAN V2. Each row corresponds to a different input image. Columns represent (a) the original input image, (b) the result obtained using the proposed method, and (c) the result produced by CycleGAN V2.
  • Figure 3: Variation of number of steps in $[2, 5, 10, 20, 40]$ with a fixed guidance scale of $1.0$.
  • Figure 4: Variation of guidance scale in the interval $(1.0, 1.9)$ with a fixed number of steps.
  • Figure : (a)