metadata

dataset_info:
  features:
    - name: caption
      dtype: string
    - name: image
      dtype: image
  splits:
    - name: train
      num_bytes: 1372144989.6
      num_examples: 3960
    - name: test
      num_bytes: 343036247.4
      num_examples: 990
  download_size: 1709055735
  dataset_size: 1715181237
language:
  - en
tags:
  - code
size_categories:
  - 100B<n<1T

Dataset Card for "BilbaoCaptions"

Dataset Description

Homepage: https://github.com/TheMrguiller/MUCSI_Modal
Repository: https://github.com/TheMrguiller/MUCSI_Modal
Paper: It is a follow up of the Flamingo model paper
Leaderboard:
Point of Contact: https://github.com/TheMrguiller/MUCSI_Modal

Dataset Summary

This dataset was collected for a proyect for a master degree in Computation and Intelligent System from University of Deusto. It was done by students and recolected from webpages famous in the Basque Country: Deia and Getimages.

Supported Tasks and Leaderboards

The dataset is prepared to used it for visual question-answering.

Languages

The dataset is in english.

Dataset Structure

Data Fields

Caption: This field has the description of the image.
Image: This field has the image corresponding to the description.

Data Splits

The dataset is split in 80% train and 20% test.

Considerations for Using the Data

The dataset has some flaws regarding to the descriptions. The descriptions sometimes are to specific for a captioning task. There are also to many futbol match data, so it isnt to well balanced. There are also some description that are to generic.

Additional Information

Dataset Curators

The curators of this dataset where the students from the Masters degree in Computation and Inteligent Systems from University of Deusto.