Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.20288 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913812106444800 |
|---|---|
| author | Saito, Shinnosuke Matsubara, Takashi |
| author_facet | Saito, Shinnosuke Matsubara, Takashi |
| contents | Diffusion models excel in content generation by implicitly learning the data manifold, yet they lack a practical method to leverage this manifold - unlike other deep generative models equipped with latent spaces. This paper introduces a novel framework that treats the data space of pre-trained diffusion models as a Riemannian manifold, with a metric derived from the score function. Experiments with MNIST and Stable Diffusion show that this geometry-aware approach yields image interpolations that are more realistic, less noisy, and more faithful to prompts than existing methods, demonstrating its potential for improved content generation and editing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_20288 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Image Interpolation with Score-based Riemannian Metrics of Diffusion Models Saito, Shinnosuke Matsubara, Takashi Computer Vision and Pattern Recognition Diffusion models excel in content generation by implicitly learning the data manifold, yet they lack a practical method to leverage this manifold - unlike other deep generative models equipped with latent spaces. This paper introduces a novel framework that treats the data space of pre-trained diffusion models as a Riemannian manifold, with a metric derived from the score function. Experiments with MNIST and Stable Diffusion show that this geometry-aware approach yields image interpolations that are more realistic, less noisy, and more faithful to prompts than existing methods, demonstrating its potential for improved content generation and editing. |
| title | Image Interpolation with Score-based Riemannian Metrics of Diffusion Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.20288 |