Toward Parking Spot Occupancy Recognition: A Self-Supervised Approach

2026-06-18

AI Summary

This study addresses the scarcity of labeled data in target domains for parking space occupancy detection by proposing a self-supervised approach that requires no annotations in the target environment. Building upon the SimCLR framework with a ResNet-50 encoder, the method introduces an innovative two-stage self-supervised pretraining and fine-tuning strategy, evaluated through leave-one-environment-out cross-validation. Experimental results demonstrate that the resulting general-purpose model achieves an average accuracy of 97.2%, which further improves to 97.8% after optimization via the proposed two-stage strategy. This performance significantly surpasses existing supervised and self-supervised baselines, marking the first achievement of highly accurate and robust parking space state recognition without any target-domain labels.

Abstract

As urban areas expand, automatic monitoring of parking lots becomes essential for efficient and sustainable cities. This work proposes a self-supervised approach for parking spot occupancy recognition that requires no labeled samples from the target parking lot. Building upon a self-supervised transfer learning fine-tuning protocol, the proposed training strategy consists of two self-supervised stages: first on unlabeled generic data and then on unlabeled target-specific data, followed by supervised fine-tuning using only generic parking lot labels. We adopt SimCLR with a ResNet-50 encoder and evaluate the method under a leave-one-out cross-environment protocol on three public datasets: PKLot, CNRPark-EXT, and PLds. We also introduce a two-stage deployment strategy in which a Strong General Model is initially deployed, followed by a Specialized Model that incorporates unlabeled images collected during the first N days of deployment in a self-supervised manner. Experimental results show that the Strong General Model alone outperforms supervised and self-supervised baselines, achieving an average accuracy of 97.2%, which further improves to 97.8% with the proposed two-stage strategy. These results demonstrate that self-supervised learning enables a scalable and labelefficient solution for real-world parking occupancy monitoring. Our trained models and source code are publicly available at https://github.com/LoanMaikon/Parking-Spot-Occupancy-Recognition.

Problem

• parking spot occupancy
• self-supervised learning
• label-efficient
• smart cities
• computer vision

Innovation

• self-supervised learning
• parking spot occupancy
• transfer learning
• label-efficient
• two-stage deployment