metadata

dataset_info:
  features:
    - name: image_id
      dtype: int64
    - name: image
      dtype: image
    - name: image_width
      dtype: int64
    - name: image_height
      dtype: int64
    - name: objects
      struct:
        - name: bbox_id
          sequence: int64
        - name: category
          sequence:
            class_label:
              names:
                '0': vehicle
                '1': truck
        - name: bbox
          sequence:
            sequence: float32
            length: 4
        - name: area
          sequence: float32
  splits:
    - name: train
      num_bytes: 951705037.8
      num_examples: 3650
    - name: test
      num_bytes: 38218268
      num_examples: 162
  download_size: 987020725
  dataset_size: 989923305.8
license: apache-2.0
pretty_name: Highway vehicles - object detection
task_categories:
  - object-detection

Dataset Card for "highway-vehicles"

This dataset contains scenes from road cameras (mainly highway cameras) with cars classified into vehicles (class 0), and trucks (class 1). Overall, it contains 3,812 scenes with 91,843 objects.

The detailed instructions are available at GitHub Repo.

Citations:

If you are using this dataset, consider citing the related paper:

Placeholder for a paper, will appear as soon as it is published :>