The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: ValueError
Message: Invalid string class label MarTA-Net@d091b237e6d207448ccacfafa3c42eea8e7553bd
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2474, in __iter__
example = _apply_feature_types_on_example(
example, self.features, token_per_repo_id=self.token_per_repo_id
)
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2391, in _apply_feature_types_on_example
encoded_example = features.encode_example(example)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 2192, in encode_example
return encode_nested_example(self, example)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1472, in encode_nested_example
{k: encode_nested_example(schema[k], obj.get(k), level=level + 1) for k in schema}
~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1495, in encode_nested_example
return schema.encode_example(obj) if obj is not None else None
~~~~~~~~~~~~~~~~~~~~~^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1168, in encode_example
example_data = self.str2int(example_data)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1105, in str2int
output = [self._strval2int(value) for value in values]
~~~~~~~~~~~~~~~~^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1126, in _strval2int
raise ValueError(f"Invalid string class label {value}")
ValueError: Invalid string class label MarTA-Net@d091b237e6d207448ccacfafa3c42eea8e7553bdNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Dataset Card for Dataset Name
This dataset contains HiRISE Mars surface images paired with segmentation masks for multi-class Martian terrain segmentation.
Dataset Details
Dataset Sources
- Repository: https://github.com/VGNET-LAB/Mars-surface-segmentation
- Paper: https://ieeexplore.ieee.org/abstract/document/11664087
Uses
Direct Use
The dataset can be used for research on multi-class semantic segmentation of Martian surface imagery, terrain-feature mapping, crater and landform analysis, and the evaluation of computer-vision models for planetary surface interpretation.
Out of scope Use
The dataset should not be used by itself to make safety-critical landing-site decisions, control spacecraft, or support operational mission decisions without independent validation and expert review. Model predictions should not be treated as definitive geological conclusions.
Dataset Structure
The supplied description reports approximately 10,000 augmented training image–mask pairs, 227 validation image–mask pairs (can be increased with augmentation as per requirement), and 200 manually annotated source images.
The 12 terrain classes and supplied color values are:
Class RGB (R, G, B)
Crater (0, 217, 251)
Bedrock (1, 176, 1)
Boulder (0, 13, 225)
Cliff (211, 2, 201)
Channel (254, 245, 6)
Dunes (254, 245, 6)
Alluvial fan (245, 0, 0)
Impact ejecta (251, 131, 1)
Ridge (113, 5, 132)
Rough terrain (143, 50, 0)
Smooth terrain (253, 172, 187)
Background (38, 38, 37)
The masks are RGB color-coded, do not load them as grayscale because this loses the class-color information. OpenCV loads color images in BGR channel order, so convert to RGB before comparing with the RGB values above. Confirm whether the uploaded mask files are color-coded or integer class-ID masks before use.
BibTeX:
@ARTICLE{11664087, author={Chatterjee, Anoushka and Kandi, Kaustubh Anand and Adhikari, Mainak}, journal={IEEE Internet of Things Journal}, title={MarTA-Net: A Terrain-Aware U-Net Enhanced with Convolutional Block Attention Modules for Martian Surface Segmentation}, year={2026}, volume={}, number={}, pages={1-1}, keywords={Modeling;Internet of Things;Dies;Training;Decoding;Modules (abstract algebra);Architecture;Computer architecture;Mars;Pixel;Semantic Segmentation;Attention U-Net;Convolutional Block Attention Mechanism;Knowledge Distillation}, doi={10.1109/JIOT.2026.3726564} }
- Downloads last month
- 43