Improve dataset card: Add paper and code links
Browse filesThis PR enhances the dataset card by adding prominent links to the paper ([STAR: A Benchmark for Astronomical Star Fields Super-Resolution](https://huggingface.co/papers/2507.16385)) and the GitHub repository (https://github.com/GuoCheng12/STAR) directly under the main title. This improves the discoverability and navigation for users interested in the research and codebase associated with the STAR dataset.
All existing valuable content, including detailed usage examples and metadata, remains unchanged.
README.md
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
---
|
| 2 |
-
license: mit
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
-
|
| 6 |
size_categories:
|
| 7 |
- 100M<n<1B
|
| 8 |
task_categories:
|
| 9 |
- image-to-image
|
|
|
|
| 10 |
tags:
|
| 11 |
- astronomy
|
| 12 |
- super-resolution
|
|
@@ -23,6 +23,8 @@ configs:
|
|
| 23 |
|
| 24 |
# STAR Dataset (Super-Resolution for Astronomical Star Fields)
|
| 25 |
|
|
|
|
|
|
|
| 26 |
The **STAR** dataset is a large-scale benchmark for developing field-level super-resolution models in astronomy. It contains **54,738 flux-consistent image pairs** derived from Hubble Space Telescope (HST) high-resolution observations and physically faithful low-resolution counterparts.
|
| 27 |
|
| 28 |
## 🌟 Key Features
|
|
@@ -97,10 +99,10 @@ Each `.npy` file contains a dictionary with the following structure:
|
|
| 97 |
### Data Fields Explanation
|
| 98 |
|
| 99 |
| Field | Description | Type | Usage |
|
| 100 |
-
|
| 101 |
-
| `image` | Raw astronomical observation data | `np.ndarray` (float32) | Main input for super-resolution
|
| 102 |
-
| `mask` | Valid pixel indicator | `np.ndarray` (bool) | Identifies accessible regions (True = valid)
|
| 103 |
-
| `attn_map` | Star finder output | `np.ndarray` (float32) | Highlights detected astronomical sources (stars, galaxies)
|
| 104 |
| `coord` | Spatial coordinates | `np.ndarray` | Position information for patch alignment |
|
| 105 |
|
| 106 |
## 💻 Usage Examples
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
language:
|
| 3 |
- en
|
| 4 |
+
license: mit
|
| 5 |
size_categories:
|
| 6 |
- 100M<n<1B
|
| 7 |
task_categories:
|
| 8 |
- image-to-image
|
| 9 |
+
pretty_name: STAR
|
| 10 |
tags:
|
| 11 |
- astronomy
|
| 12 |
- super-resolution
|
|
|
|
| 23 |
|
| 24 |
# STAR Dataset (Super-Resolution for Astronomical Star Fields)
|
| 25 |
|
| 26 |
+
[Paper](https://huggingface.co/papers/2507.16385) | [Code](https://github.com/GuoCheng12/STAR)
|
| 27 |
+
|
| 28 |
The **STAR** dataset is a large-scale benchmark for developing field-level super-resolution models in astronomy. It contains **54,738 flux-consistent image pairs** derived from Hubble Space Telescope (HST) high-resolution observations and physically faithful low-resolution counterparts.
|
| 29 |
|
| 30 |
## 🌟 Key Features
|
|
|
|
| 99 |
### Data Fields Explanation
|
| 100 |
|
| 101 |
| Field | Description | Type | Usage |
|
| 102 |
+
|-------|-------------|------|-------|\
|
| 103 |
+
| `image` | Raw astronomical observation data | `np.ndarray` (float32) | Main input for super-resolution |\
|
| 104 |
+
| `mask` | Valid pixel indicator | `np.ndarray` (bool) | Identifies accessible regions (True = valid) |\
|
| 105 |
+
| `attn_map` | Star finder output | `np.ndarray` (float32) | Highlights detected astronomical sources (stars, galaxies) |\
|
| 106 |
| `coord` | Spatial coordinates | `np.ndarray` | Position information for patch alignment |
|
| 107 |
|
| 108 |
## 💻 Usage Examples
|