Datasets:
Add FAME segmentation target overview to dataset card
Browse files
README.md
CHANGED
|
@@ -29,9 +29,17 @@ If you use this benchmark, please cite the FAME paper. The BibTeX entry is provi
|
|
| 29 |
|
| 30 |
This Hugging Face dataset contains the released FAME JSON/JSONL protocols plus all referenced image and mask pixels. To keep the repository usable on Hugging Face, the protocols are packaged in `protocols/FAME_benchmark_protocols.tar.gz`, and the pixels are stored as Parquet shards with `image` and `mask` columns so samples can be inspected in Dataset Viewer/Data Studio. After extracting the protocol archive, `tools/extract_pixels_from_parquet.py` can materialize the raw image and mask files referenced by the protocol JSONL files.
|
| 31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
## Directory Layout
|
| 33 |
|
| 34 |
```text
|
|
|
|
|
|
|
| 35 |
protocols/
|
| 36 |
FAME_benchmark_protocols.tar.gz
|
| 37 |
|
|
|
|
| 29 |
|
| 30 |
This Hugging Face dataset contains the released FAME JSON/JSONL protocols plus all referenced image and mask pixels. To keep the repository usable on Hugging Face, the protocols are packaged in `protocols/FAME_benchmark_protocols.tar.gz`, and the pixels are stored as Parquet shards with `image` and `mask` columns so samples can be inspected in Dataset Viewer/Data Studio. After extracting the protocol archive, `tools/extract_pixels_from_parquet.py` can materialize the raw image and mask files referenced by the protocol JSONL files.
|
| 31 |
|
| 32 |
+
## What Is Segmented?
|
| 33 |
+
|
| 34 |
+
The montage below shows representative positive target masks from the released FAME tasks. Magenta indicates the foreground mask and yellow indicates the mask contour.
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
|
| 38 |
## Directory Layout
|
| 39 |
|
| 40 |
```text
|
| 41 |
+
assets/
|
| 42 |
+
fame_segmentation_targets_overview.png
|
| 43 |
protocols/
|
| 44 |
FAME_benchmark_protocols.tar.gz
|
| 45 |
|