Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -11,14 +11,14 @@ license: cc-by-4.0
|
|
| 11 |
> *VoDaSuRe: A Large-Scale Dataset Revealing Domain Shift in Volumetric Super-Resolution*
|
| 12 |
|
| 13 |
The dataset consists of **32 volumetric scans of 16 samples**, each acquired under varying imaging conditions, enabling research on generalization, robustness, and cross-domain learning in 3D super-resolution.
|
| 14 |
-
|
| 15 |
|
| 16 |
## 🔗 Resources
|
| 17 |
|
| 18 |
* **Project page**: https://augusthoeg.github.io/VoDaSuRe/
|
| 19 |
* **Paper (arXiv)**: https://arxiv.org/abs/2603.23153
|
| 20 |
* **Code & pipelines**: https://github.com/AugustHoeg/VoxelSR
|
| 21 |
-
|
| 22 |
|
| 23 |
## Dataset Structure
|
| 24 |
|
|
@@ -31,8 +31,8 @@ VoDaSuRe/
|
|
| 31 |
└── test/
|
| 32 |
```
|
| 33 |
|
| 34 |
-
Each split contains volumetric data stored in OME-Zarr format, a hierarchical and chunked format that enables efficient, lazy loading of large-scale volumetric data.
|
| 35 |
-
|
| 36 |
|
| 37 |
## Data Format (OME-Zarr)
|
| 38 |
|
|
@@ -62,7 +62,6 @@ ome.zarr
|
|
| 62 |
* **HR**: High-resolution reference volumes
|
| 63 |
* **LR**: Low-resolution volumes (unregistered)
|
| 64 |
* **REG**: Registered and intensity-matched low-resolution volumes
|
| 65 |
-
---
|
| 66 |
|
| 67 |
## Dataset Size
|
| 68 |
|
|
@@ -70,7 +69,6 @@ ome.zarr
|
|
| 70 |
* **Disk requirement after extraction**: ~500 GB
|
| 71 |
|
| 72 |
⚠️ Ensure sufficient disk space before downloading.
|
| 73 |
-
---
|
| 74 |
|
| 75 |
## Download Instructions
|
| 76 |
|
|
@@ -95,7 +93,6 @@ snapshot_download(
|
|
| 95 |
git lfs install
|
| 96 |
git clone https://huggingface.co/datasets/AugustHoeg/VoDaSuRe
|
| 97 |
```
|
| 98 |
-
---
|
| 99 |
|
| 100 |
## Data Usage
|
| 101 |
|
|
@@ -112,7 +109,6 @@ After extraction, the dataset can be accessed using libraries supporting OME-Zar
|
|
| 112 |
* `zarr`
|
| 113 |
* `ome-zarr-py`
|
| 114 |
* `dask`
|
| 115 |
-
---
|
| 116 |
|
| 117 |
## Intended Use
|
| 118 |
|
|
@@ -121,14 +117,12 @@ VoDaSuRe is designed for:
|
|
| 121 |
* Volumetric super-resolution (3D SR)
|
| 122 |
* Domain generalization and domain shift analysis
|
| 123 |
* Benchmarking learning-based SR methods under realistic acquisition scenarios
|
| 124 |
-
---
|
| 125 |
|
| 126 |
## Dataset Creation
|
| 127 |
|
| 128 |
The dataset was created using **laboratory CT (Lab-CT) imaging systems**, capturing paired high- and low-resolution volumetric scans under varying acquisition conditions.
|
| 129 |
|
| 130 |
Further details are available in the associated paper and project page.
|
| 131 |
-
---
|
| 132 |
|
| 133 |
## Citation
|
| 134 |
|
|
@@ -143,7 +137,6 @@ If you use this dataset, please cite our paper:
|
|
| 143 |
url={https://augusthoeg.github.io/VoDaSuRe/}
|
| 144 |
}
|
| 145 |
```
|
| 146 |
-
---
|
| 147 |
|
| 148 |
## Contact
|
| 149 |
|
|
|
|
| 11 |
> *VoDaSuRe: A Large-Scale Dataset Revealing Domain Shift in Volumetric Super-Resolution*
|
| 12 |
|
| 13 |
The dataset consists of **32 volumetric scans of 16 samples**, each acquired under varying imaging conditions, enabling research on generalization, robustness, and cross-domain learning in 3D super-resolution.
|
| 14 |
+
|
| 15 |
|
| 16 |
## 🔗 Resources
|
| 17 |
|
| 18 |
* **Project page**: https://augusthoeg.github.io/VoDaSuRe/
|
| 19 |
* **Paper (arXiv)**: https://arxiv.org/abs/2603.23153
|
| 20 |
* **Code & pipelines**: https://github.com/AugustHoeg/VoxelSR
|
| 21 |
+
|
| 22 |
|
| 23 |
## Dataset Structure
|
| 24 |
|
|
|
|
| 31 |
└── test/
|
| 32 |
```
|
| 33 |
|
| 34 |
+
Each split contains volumetric data stored in **OME-Zarr** format, a hierarchical and chunked format that enables efficient, lazy loading of large-scale volumetric data.
|
| 35 |
+
|
| 36 |
|
| 37 |
## Data Format (OME-Zarr)
|
| 38 |
|
|
|
|
| 62 |
* **HR**: High-resolution reference volumes
|
| 63 |
* **LR**: Low-resolution volumes (unregistered)
|
| 64 |
* **REG**: Registered and intensity-matched low-resolution volumes
|
|
|
|
| 65 |
|
| 66 |
## Dataset Size
|
| 67 |
|
|
|
|
| 69 |
* **Disk requirement after extraction**: ~500 GB
|
| 70 |
|
| 71 |
⚠️ Ensure sufficient disk space before downloading.
|
|
|
|
| 72 |
|
| 73 |
## Download Instructions
|
| 74 |
|
|
|
|
| 93 |
git lfs install
|
| 94 |
git clone https://huggingface.co/datasets/AugustHoeg/VoDaSuRe
|
| 95 |
```
|
|
|
|
| 96 |
|
| 97 |
## Data Usage
|
| 98 |
|
|
|
|
| 109 |
* `zarr`
|
| 110 |
* `ome-zarr-py`
|
| 111 |
* `dask`
|
|
|
|
| 112 |
|
| 113 |
## Intended Use
|
| 114 |
|
|
|
|
| 117 |
* Volumetric super-resolution (3D SR)
|
| 118 |
* Domain generalization and domain shift analysis
|
| 119 |
* Benchmarking learning-based SR methods under realistic acquisition scenarios
|
|
|
|
| 120 |
|
| 121 |
## Dataset Creation
|
| 122 |
|
| 123 |
The dataset was created using **laboratory CT (Lab-CT) imaging systems**, capturing paired high- and low-resolution volumetric scans under varying acquisition conditions.
|
| 124 |
|
| 125 |
Further details are available in the associated paper and project page.
|
|
|
|
| 126 |
|
| 127 |
## Citation
|
| 128 |
|
|
|
|
| 137 |
url={https://augusthoeg.github.io/VoDaSuRe/}
|
| 138 |
}
|
| 139 |
```
|
|
|
|
| 140 |
|
| 141 |
## Contact
|
| 142 |
|