canada-guesser/Canadian-streetview-cities

Canadian Street View Cities Dataset

Overview

A street-view image dataset created to train and evaluate models for city-level image classification across major Canadian cities. Each entry includes an image and its corresponding city label.

Purpose

The dataset is intended for building models that recognize the Canadian city in which a street-view scene was captured.

Data Source

All images were collected from Mapillary, using geographic bounding boxes for each target city.
Collection was performed individually, collecting 10 000 images for each city.

Dataset Structure

• Train split: 135,000 images
• Test split: 15,000 images
• Data is stored in Parquet files
• Includes 15 Canadian cities, represented in the label and city fields

Features

• image — street-view image
• label — integer class ID (0–14)
• city — text name of the city

Preprocessing

No preprocessing was applied. Images are provided exactly as downloaded from Mapillary.

Known Limitations

• Some cities, especially Saskatoon and Halifax, contain dashcam-style captures including visible dashboards.
  These can influence model behavior if not processed correctly or accounted for.

Example Usage

from datasets import load_dataset

dataset = load_dataset("canada-guesser/Canadian-streetview-cities", streaming=False)
train_ds = dataset["train"]
test_ds = dataset["test"]

Original Dataset: https://huggingface.co/datasets/SABR22/Canadian-streetview-cities

Citation

If you use this dataset or models, please cite:

1. Stephen Rebel, Danial McIntyre, Sharav Bali. Canadian Street View Classifier. Hugging Face, 2025.