S-NODE: Explicit and Reversible Image Translation Encoding With Neural ODEs
Lists
Only you see this
Abstract
Score-based diffusion models achieve high-quality data generation through an iterative denoising process. However, the stochastic term in the diffusion process prevents them from accomplishing reversible generative modeling. To tackle this problem, we present S-NODE, a novel generative model capable of reversible and conditional data generation. Unlike score-based models, S-NODE is an entirely deterministic generative method bridging the score function and neural ordinary differential equations (ODEs). Firstly, we propose and prove an ODE utilizing a score-related difference as the drift term to model transformations between two certain data distributions. Secondly, we suggest a path-constrained loss to reduce truncation errors, enhancing the model’s capabilities in generating high-quality samples. Thirdly, S-NODE can use a single conditional model to generate and translate cross-class images in all stages without additional training. Extensive experiments on various tasks demonstrate the effectiveness and reversibility of our method. Compared to other ODE-based and score-based methods, S-NODE achieves superior performance (FID of 2.29 & IS of 9.96) on CIFAR-10 and facilitates reversible image translation and image interpolation on CelebA, MetFace, and AFHQ datasets. Code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Tocmac/S-NODE</uri>.
BibTeX
@article{wang2026s,
author = {Xu Wang and Dong Pang and Zhiyuan You and Xinping Guan and Xinyi Le},
title = {{S-NODE: Explicit and Reversible Image Translation Encoding With Neural ODEs}},
journal = {IEEE Trans. Artif. Intell.},
year = {2026},
doi = {10.1109/tai.2025.3619457},
}Loading related content...