Image-Based Parking Space Occupancy Classification: Dataset and Baseline

TL;DR

This paper presents a new, systematically annotated dataset for parking space occupancy classification, along with a baseline model that achieves high accuracy and outperforms existing methods.

Contribution

Introduction of the ACPDS dataset with unique views and annotations, and a baseline model demonstrating strong performance.

Findings

Achieved 98% accuracy on unseen parking lots.

Outperformed existing models significantly.

Shared dataset, code, and models openly.

Abstract

We introduce a new dataset for image-based parking space occupancy classification: ACPDS. Unlike in prior datasets, each image is taken from a unique view, systematically annotated, and the parking lots in the train, validation, and test sets are unique. We use this dataset to propose a simple baseline model for parking space occupancy classification, which achieves 98% accuracy on unseen parking lots, significantly outperforming existing models. We share our dataset, code, and trained models under the MIT license.