Vehicle Occurrence-based Parking Space Detection

TL;DR

This paper introduces an automatic parking space detection method using instance segmentation and vehicle occurrence heat maps, achieving high accuracy without manual labeling of parking spots.

Contribution

It presents a novel automated approach for parking space detection that eliminates the need for manual spot labeling, leveraging vehicle occurrence data.

Findings

Achieved AP25 score up to 95.60%

Achieved AP50 score up to 79.90%

Validated on multiple datasets with strong results

Abstract

Smart-parking solutions use sensors, cameras, and data analysis to improve parking efficiency and reduce traffic congestion. Computer vision-based methods have been used extensively in recent years to tackle the problem of parking lot management, but most of the works assume that the parking spots are manually labeled, impacting the cost and feasibility of deployment. To fill this gap, this work presents an automatic parking space detection method, which receives a sequence of images of a parking lot and returns a list of coordinates identifying the detected parking spaces. The proposed method employs instance segmentation to identify cars and, using vehicle occurrence, generate a heat map of parking spaces. The results using twelve different subsets from the PKLot and CNRPark-EXT parking lot datasets show that the method achieved an AP25 score up to 95.60\% and AP50 score up to 79.90\%.