Smoker Classification Dataset

Table of Contents

• Overview
• Dataset Structure
• Reference
• License

Overview

This dataset contains images of people in everyday settings — including cafés, streets, workplaces, and social gatherings — captured from various angles and under different lighting conditions. Each image depicts one or more individuals and is labeled to indicate whether smoking activity is present.

The dataset is designed for the Smoker Classification Challenge, enabling participants to build and evaluate binary image classification models for detecting smoking behavior.

Property	Value
Total samples	940
Training samples	716
Test samples	224
Classes	2 (0: Non-Smoker, 1: Smoker)

Dataset Structure

Clone the repository:

git clone [email protected]:AIOZAI/smoker_classification_dataset.git

Or download directly: Smoker Classification Data

Files

Folder	Samples	Description
data/train/	716 Images	Images with labels in ground_truth.csv
data/test/	224 Images	Images reserved for public scoring and evaluation

Data Fields

The ground_truth.csv file contains the labels for the training images:

Field	Type	Description
id	string	The image filename.
label	integer	The target class (0: Non-Smoker, 1: Smoker)

Example: ground_truth.csv

id,label
6138699878db49328e6f59e6337cd7ee.jpg,0
7d16690af8d0420f9d6b9647354f0300.jpg,0

Reference

For more information about the dataset, please visit the Mendeley Data page.

License

This dataset is released under the Creative Commons Attribution 4.0 International License (CC BY 4.0).