README.md

metadata

dataset_info:
  features:
    - name: index
      dtype: int64
    - name: video_index
      dtype: string
    - name: question
      dtype: string
    - name: answer
      dtype: string
    - name: eval_method
      dtype: string
    - name: option
      sequence: string
    - name: task_type
      dtype: string
    - name: task
      dtype: string
    - name: duration
      dtype: float64
    - name: meta
      struct:
        - name: fps
          dtype: float64
        - name: frame_count
          dtype: int64
  splits:
    - name: train
      num_bytes: 500262
      num_examples: 2000
  download_size: 167816
  dataset_size: 500262
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
task_categories:
  - video-text-to-text

This dataset is from the paper MME-VideoOCR: Evaluating OCR-Based Capabilities of Multimodal LLMs in Video Scenarios.

See https://github.com/DogNeverSleep/MME-VideoOCR_Dataset for more information, including the license.