Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning
Paper • 2003.02437 • Published
Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: ValueError
Message: Invalid string class label DroneVehicle@b913e4d5ab34b0deff02b94b7a4c52a0b7175350
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 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2386, 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 2303, 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 2178, in encode_example
return encode_nested_example(self, example)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1460, 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 1483, 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 1158, in encode_example
example_data = self.str2int(example_data)
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1095, in str2int
output = [self._strval2int(value) for value in values]
~~~~~~~~~~~~~~~~^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/features/features.py", line 1116, in _strval2int
raise ValueError(f"Invalid string class label {value}")
ValueError: Invalid string class label DroneVehicle@b913e4d5ab34b0deff02b94b7a4c52a0b7175350Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
VisDrone-DroneVehicle
Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning (paper).
Abstract
Drone-based vehicle detection aims at finding the vehicle locations and categories in an aerial image. It empowers smart city traffic management and disaster rescue. Researchers have made mount of efforts in this area and achieved considerable progress. Nevertheless, it is still a challenge when the objects are hard to distinguish, especially in low light conditions. To tackle this problem, we construct a large-scale drone-based RGB-Infrared vehicle detection dataset, termed DroneVehicle. Our DroneVehicle collects 28,439 RGB-Infrared image pairs, covering urban roads, residential areas, parking lots, and other scenarios from day to night. Due to the great gap between RGB and infrared images, cross-modal images provide both effective information and redundant information. To address this dilemma, we further propose an uncertainty-aware cross-modality vehicle detection (UA-CMDet) framework to extract complementary information from cross-modal images, which can significantly improve the detection performance in low light conditions. An uncertainty-aware module (UAM) is designed to quantify the uncertainty weights of each modality, which is calculated by the cross-modal Intersection over Union (IoU) and the RGB illumination value. Furthermore, we design an illumination-aware cross-modal non-maximum suppression algorithm to better integrate the modal-specific information in the inference phase. Extensive experiments on the DroneVehicle dataset demonstrate the flexibility and effectiveness of the proposed method for cross-modality vehicle detection.
Dataset
The DroneVehicle dataset consists of a total of 56,878 images collected by the drone, half of which are RGB images, and the resting are infrared images. We have made rich annotations with oriented bounding boxes for the five categories. Among them, car has 389,779 annotations in RGB images, and 428,086 annotations in infrared images, truck has 22,123 annotations in RGB images, and 25,960 annotations in infrared images, bus has 15,333 annotations in RGB images, and 16,590 annotations in infrared images, van has 11,935 annotations in RGB images, and 12,708 annotations in infrared images, and freight car has 13,400 annotations in RGB images, and 17,173 annotations in infrared image. This dataset is available on the download page.
In DroneVehicle, to annotate the objects at the image boundaries, we set a white border with a width of 100 pixels on the top, bottom, left and right of each image, so that the downloaded image scale is 840 x 712. When training our detection network, we can perform pre-processing to remove the surrounding white border and change the image scale to 640 x 512.
- Downloads last month
- 39