Add metadata and links
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,5 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Discrete Diffusion Bridges for Spatiotemporally Aligned Image Translation and Generation
|
| 2 |
|
|
|
|
|
|
|
| 3 |
|
| 4 |
## 🌟 Overview
|
| 5 |
|
|
@@ -8,4 +15,4 @@
|
|
| 8 |
- A **hybrid absorption mechanism** that mixes source-image and mask tokens, preserving source structure as spatial anchors.
|
| 9 |
- An **information-guided noise schedule** that aligns training corruption with the easy-first decoding process used at inference time.
|
| 10 |
|
| 11 |
-
DDB supports text-guided image editing, structural image translation, and text-to-image generation.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pipeline_tag: image-to-image
|
| 3 |
+
library_name: diffusers
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
# Discrete Diffusion Bridges for Spatiotemporally Aligned Image Translation and Generation
|
| 7 |
|
| 8 |
+
**Paper**: [Discrete Diffusion Bridges for Spatiotemporally Aligned Image Translation and Generation](https://huggingface.co/papers/2608.29997)
|
| 9 |
+
**Code**: [GitHub](https://github.com/HKU-HealthAI/DDB)
|
| 10 |
|
| 11 |
## 🌟 Overview
|
| 12 |
|
|
|
|
| 15 |
- A **hybrid absorption mechanism** that mixes source-image and mask tokens, preserving source structure as spatial anchors.
|
| 16 |
- An **information-guided noise schedule** that aligns training corruption with the easy-first decoding process used at inference time.
|
| 17 |
|
| 18 |
+
DDB supports text-guided image editing, structural image translation, and text-to-image generation.
|