# An Ensemble-Based Self-Taught Learning Approach for Parking Space Classification Under Limited Data

arXiv: https://arxiv.org/abs/2609.03258 · HTML: https://mlsanity.com/paper/2609.03258/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 16 references that are in the map.
- First posted on arXiv 3 Sep 2026.

## Builds on (most cited first, 6 of 6)

- [A Systematic Review on Computer Vision-Based Parking Lot Management Applied on Public Datasets](https://mlsanity.com/paper/2203.06463/) (2022)
- [Evaluation of Different Annotation Strategies for Deployment of Parking Spaces Classification Systems](https://mlsanity.com/paper/2207.11372/) (2022)
- [Vehicle Occurrence-based Parking Space Detection](https://mlsanity.com/paper/2306.09940/) (2023)
- [Deep Single Models vs. Ensembles: Insights for a Fast Deployment of Parking Monitoring Systems](https://mlsanity.com/paper/2309.16495/) (2023)
- [Optimizing Parking Space Classification: Distilling Ensembles into Lightweight Classifiers](https://mlsanity.com/paper/2410.14705/) (2024)
- [On the benefits of self-taught learning for brain decoding](https://mlsanity.com/paper/2209.10099/) (2022)

## Abstract (arXiv)

Parking spot classification is a fundamental task in intelligent transportation systems, yet most deep learning approaches rely on large amounts of annotated data and exhibit limited generalization across heterogeneous environments. To address these limitations, we investigate a self-taught learning framework based on unsupervised representation learning with convolutional autoencoders. The proposed approach learns transferable visual representations from unlabeled data and reuses the learned encoders as fixed feature extractors for supervised classification with limited annotated samples in the target domain. To further enhance robustness and mitigate architectural bias, an ensemble of heterogeneous autoencoders is employed, with independent classifier heads and prediction fusion at inference time. Experiments conducted on the PKLot and CNRPark benchmarks under cross-dataset evaluation protocols show that the proposed ensemble-based strategy substantially reduces annotation requirements while improving robustness under significant domain shifts, achieving accuracies between 93% and 96% in data-constrained scenarios.