Datasets:

ntucllab
/

aclcifar20

images imagewidth (px) 32 32	ord_labels class label 20 classes	cl_labels sequencelengths 3 3
	16small mammals	[ 1, 4, 5 ]
	15reptiles	[ 2, 14, 4 ]
	15reptiles	[ 3, 19, 6 ]
	2flowers	[ 13, 13, 15 ]
	14people	[ 6, 19, 1 ]
	18transportation vehicles	[ 4, 1, 11 ]
	4fruit, vegetables and mushrooms	[ 17, 19, 14 ]
	2flowers	[ 15, 11, 9 ]
	17trees	[ 6, 6, 5 ]
	14people	[ 3, 13, 6 ]
	19non-transportation vehicles	[ 0, 8, 13 ]
	9large man-made outdoor things	[ 4, 7, 11 ]
	19non-transportation vehicles	[ 14, 19, 1 ]
	10large natural outdoor scenes	[ 6, 13, 6 ]
	6household furniture	[ 13, 14, 7 ]
	14people	[ 17, 19, 15 ]
	5household electrical devices	[ 6, 7, 17 ]
	0aquatic_mammals	[ 8, 7, 6 ]
	14people	[ 6, 8, 17 ]
	7insects	[ 3, 6, 6 ]
	6household furniture	[ 4, 11, 15 ]
	13non-insect invertebrates	[ 3, 6, 5 ]
	2flowers	[ 4, 0, 15 ]
	0aquatic_mammals	[ 4, 5, 6 ]
	12medium-sized mammals	[ 19, 7, 6 ]
	18transportation vehicles	[ 13, 11, 2 ]
	2flowers	[ 7, 15, 0 ]
	16small mammals	[ 15, 9, 6 ]
	11large omnivores and herbivores	[ 17, 5, 7 ]
	12medium-sized mammals	[ 15, 17, 19 ]
	9large man-made outdoor things	[ 7, 0, 2 ]
	5household electrical devices	[ 2, 15, 1 ]
	8large carnivores and bear	[ 6, 4, 5 ]
	16small mammals	[ 1, 15, 6 ]
	19non-transportation vehicles	[ 6, 13, 2 ]
	10large natural outdoor scenes	[ 7, 16, 3 ]
	17trees	[ 1, 13, 6 ]
	2flowers	[ 1, 12, 11 ]
	5household electrical devices	[ 19, 4, 5 ]
	12medium-sized mammals	[ 5, 5, 17 ]
	12medium-sized mammals	[ 5, 1, 18 ]
	2flowers	[ 14, 8, 19 ]
	17trees	[ 8, 8, 4 ]
	16small mammals	[ 2, 3, 4 ]
	9large man-made outdoor things	[ 4, 15, 14 ]
	0aquatic_mammals	[ 6, 11, 17 ]
	13non-insect invertebrates	[ 3, 2, 18 ]
	8large carnivores and bear	[ 7, 6, 7 ]
	12medium-sized mammals	[ 17, 8, 17 ]
	0aquatic_mammals	[ 17, 19, 5 ]
	15reptiles	[ 9, 3, 2 ]
	4fruit, vegetables and mushrooms	[ 5, 8, 0 ]
	18transportation vehicles	[ 7, 6, 15 ]
	0aquatic_mammals	[ 1, 4, 13 ]
	13non-insect invertebrates	[ 1, 6, 15 ]
	5household electrical devices	[ 1, 13, 17 ]
	17trees	[ 7, 15, 1 ]
	2flowers	[ 1, 8, 5 ]
	5household electrical devices	[ 4, 6, 7 ]
	10large natural outdoor scenes	[ 6, 8, 6 ]
	18transportation vehicles	[ 2, 4, 2 ]
	18transportation vehicles	[ 13, 13, 13 ]
	18transportation vehicles	[ 6, 0, 6 ]
	18transportation vehicles	[ 0, 2, 7 ]
	0aquatic_mammals	[ 6, 6, 19 ]
	17trees	[ 3, 4, 0 ]
	2flowers	[ 1, 15, 6 ]
	10large natural outdoor scenes	[ 19, 17, 6 ]
	13non-insect invertebrates	[ 9, 17, 6 ]
	13non-insect invertebrates	[ 3, 11, 9 ]
	15reptiles	[ 3, 19, 18 ]
	7insects	[ 15, 1, 13 ]
	9large man-made outdoor things	[ 11, 2, 2 ]
	9large man-made outdoor things	[ 1, 13, 2 ]
	13non-insect invertebrates	[ 18, 14, 2 ]
	3food_containers	[ 1, 4, 17 ]
	12medium-sized mammals	[ 17, 6, 18 ]
	19non-transportation vehicles	[ 3, 14, 6 ]
	9large man-made outdoor things	[ 7, 1, 7 ]
	16small mammals	[ 2, 2, 5 ]
	14people	[ 0, 2, 6 ]
	17trees	[ 4, 8, 13 ]
	6household furniture	[ 7, 4, 19 ]
	6household furniture	[ 0, 16, 9 ]
	16small mammals	[ 6, 5, 11 ]
	2flowers	[ 8, 1, 3 ]
	15reptiles	[ 2, 2, 17 ]
	10large natural outdoor scenes	[ 15, 14, 6 ]
	15reptiles	[ 2, 19, 19 ]
	17trees	[ 0, 2, 8 ]
	1fish	[ 18, 8, 0 ]
	12medium-sized mammals	[ 1, 6, 2 ]
	0aquatic_mammals	[ 4, 2, 1 ]
	19non-transportation vehicles	[ 4, 15, 15 ]
	1fish	[ 13, 17, 5 ]
	9large man-made outdoor things	[ 14, 8, 13 ]
	16small mammals	[ 13, 14, 3 ]
	11large omnivores and herbivores	[ 0, 13, 13 ]
	7insects	[ 18, 1, 7 ]
	17trees	[ 9, 13, 7 ]

End of preview. Expand in Data Studio

Dataset Card for ACLCIFAR20

This Complementary labeled CIFAR20 dataset contains auto-labeled complementary labels for all 50000 images in the training split of CIFAR20. We group 4-6 categories as a superclass and collect the complementary labels of these 20 superclasses.

For more details, please visit our github or paper.

Dataset Structure

Data Instances

A sample from the training set is provided below:

{
    'images': <PIL.PngImagePlugin.PngImageFile image mode=RGB size=32x32 at 0x7BFA04799E10>, 
    'ord_labels': 16, 
    'cl_labels': [1, 4, 5]
}

Data Fields

images: A PIL.Image.Image object containing the 32x32 image.
ord_labels: The ordinary labels of the images, and they are labeled from 0 to 19 as follows:

0: aquatic_mammals 1: fish 2: flowers 3: food_containers 4: fruit, vegetables and mushrooms 5: household electrical devices 6: household furniture 7: insects 8: large carnivores and bear 9: large man-made outdoor things 10: large natural outdoor scenes 11: large omnivores and herbivores 12: medium-sized mammals 13: non-insect invertebrates 14: people 15: reptiles 16: small mammals 17: trees 18: transportation vehicles 19: non-transportation vehicles
cl_labels: Three complementary labels for each image.

Citing

If you find this dataset useful, please cite the following:

@InProceedings{aclimage2025,
  author="Mai, Tan-Ha and Ye, Nai-Xuan and Kuan, Yu-Wei and Lu, Po-Yi and Lin, Hsuan-Tien",
  title="The Unexplored Potential of Vision-Language Models for Generating Large-Scale Complementary-Label Learning Data",
  booktitle="Pacific-Asia Conference on Knowledge Discovery and Data Mining",
  year="2025",
  pages="90--102"
}

Downloads last month: 19