Datasets:

makeabilitylab
/

disabilityparking

license: mit
task_categories:
  - object-detection
tags:
  - disability-parking
  - accessibility
  - streetscape
dataset_info:
  features:
    - name: image
      dtype: image
    - name: width
      dtype: int32
    - name: height
      dtype: int32
    - name: objects
      sequence:
        - name: bbox
          sequence: float32
          length: 4
        - name: category
          dtype: int64
        - name: area
          dtype: float32
        - name: iscrowd
          dtype: bool
        - name: id
          dtype: int64
        - name: segmentation
          sequence:
            sequence: float32
  splits:
    - name: train
      num_examples: 3688
    - name: test
      num_examples: 717
    - name: validation
      num_examples: 720

AccessParkCV

AccessParkCV is a deep learning pipeline that detects and characterizes the width of disability parking spaces from orthorectified aerial imagery. We publish a dataset of 7,069 labeled parking spaces (and 4,693 labeled access aisles), which we used to train the models making AccessParkCV possible.

(This repo contains the data in a HuggingFace format. For raw COCO format, see link).

Dataset Description

This is an object detection dataset with 8 classes:

objects
access_aisle
curbside
dp_no_aisle
dp_one_aisle
dp_two_aisle
one_aisle
two_aisle

Dataset Structure

Data Fields

image: PIL Image object
width: Image width in pixels
height: Image height in pixels
objects: Dictionary containing:
- bbox: List of bounding boxes in [x_min, y_min, x_max, y_max] format
- category: List of category IDs
- area: List of bounding box areas
- iscrowd: List of crowd flags (boolean)
- id: List of annotation IDs
- segmentation: List of polygon segmentations (each as list of [x1,y1,x2,y2,...] coordinates)

Category IDs to Category

Category ID	Class
0	objects
1	access_aisle
2	curbside
3	dp_no_aisle
4	dp_one_aisle
5	dp_two_aisle
6	one_aisle
7	two_aisle

Data Sources

Region	Lat/Long Bounding Coordinates	Source Resolution	# images in dataset
Seattle	(47.9572, -122.4489), (47.4091, -122.1551)	3 inch/pixel	2,790
Washington D.C.	(38.9979, -77.1179), (38.7962, -76.9008)	3 inch/pixel	1,801
Spring Hill	(35.7943, -87.0034), (35.6489, -86.8447)	Unknown	534
Total			5,125

Class Composition

Class	Quantity in dataset
access_aisle	4,693
curbside	36
dp_no_aisle	300
dp_one_aisle	2,790
dp_two_aisle	402
one_aisle	3,424
two_aisle	117
Total	11,762

Data Splits

Split	Examples
train	3688
test	717
valid	720

Class splits

Usage

from datasets import load_dataset

train_dataset = load_dataset("makeabilitylab/disabilityparking", split="train", streaming=True)

example = next(iter(train_dataset))

# Example of accessing an item
image = example["image"]
bboxes = example["objects"]["bbox"]
categories = example["objects"]["category"]
segmentations = example["objects"]["segmentation"]  # Polygon coordinates

Citation

@inproceedings{hwang_wherecanIpark,
  title={Where Can I Park? Understanding Human Perspectives and Scalably Detecting Disability Parking from Aerial Imagery},
  author={Hwang, Jared and Li, Chu and Kang, Hanbyul and Hosseini, Maryam and Froehlich, Jon E.},
  booktitle={The 27th International ACM SIGACCESS Conference on Computers and Accessibility},
  series={ASSETS '25},
  pages={20 pages},
  year={2025},
  month={October},
  address={Denver, CO, USA},
  publisher={ACM},
  location={New York, NY, USA},
  doi={10.1145/3663547.3746377},
  url={https://doi.org/10.1145/3663547.3746377}
}