Browse the VizWiz Dataset

Search

Within visual question

Within answers to visual question

Within captions

Image by filename

Filter

Reasons why answers differ: Skills needed by an AI system to automatically answer the visual question

TXT - Text recognition
OBJ - Object recognition
COL - Color recognition
CNT - Counting
OTH - Any other skill

Quality issues in the image

NONE - No issues
BLR - Image is blurry
BRT - Image is too bright
DRK - Image is too dark
OBS - Camera is obstructed (e.g. by finger)
FRM - Image is ill-framed
ROT - Image needs to be rotated
OTH - Any other issue

Text presence in the image

Show images with text

Show images without text

Show all images

View

Visual question with answers

Reasons why answers differ

Captions

Skills

Image quality issues

Text presence

Sort results by

If you use this VizWiz Dataset Browser in a publication, we would appreciate citations to the following paper:

Bhattacharya, N., & Gurari, D. (2019). VizWiz Dataset Browser: A Tool for Visualizing Machine Learning Datasets. arXiv preprint arXiv:1912.09336.

Bibtex entry:

@article{vizwiz_browser,
    title={VizWiz Dataset Browser: A Tool for Visualizing Machine Learning Datasets},
    author={Bhattacharya, Nilavra and Gurari, Danna},
    journal={arXiv preprint arXiv:1912.09336},
    year={2019}
}

Welcome! Here you can:
- search for words and phrases within the visual questions and/or answers
- search for specific image(s) by filename
- search visual questions based on reasons that its answers differ
- search for text within captions
- view VQAs for selected skills needed by AI systems to automatically answer the Visual Question
- view VQAs for selected Quality Issues
- view images based on whether they contain text
- or simply choose to See 50 Random Images with Annotations