README.md

metadata

dataset_info:
  features:
    - name: image_id
      dtype: int64
    - name: filename
      dtype: string
    - name: image
      dtype: image
    - name: height
      dtype: int32
    - name: width
      dtype: int32
    - name: objects
      sequence:
        - name: id
          dtype: int64
        - name: category
          dtype:
            class_label:
              names:
                '0': traffic sign
                '1': motorcycle
                '2': car
                '3': rider
                '4': person
                '5': truck
                '6': autorickshaw
                '7': vehicle fallback
                '8': bus
        - name: bbox
          sequence: float32
          length: 4
        - name: area
          dtype: float32
  splits:
    - name: train
      num_bytes: 16537488347.988
      num_examples: 31124
    - name: validation
      num_bytes: 5109718001.124417
      num_examples: 10171
  download_size: 21636854209
  dataset_size: 21647206349.11242
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
task_categories:
  - object-detection
size_categories:
  - 10K<n<100K

The IDD Object Detection dataset containing 40K images with CPPE-5 like (or YOLO) dataset annotation format.
Refer to the original dataset: https://idd.insaan.iiit.ac.in