PPD: A New Valet Parking Pedestrian Fisheye Dataset for Autonomous Driving

TL;DR

This paper introduces the Parking Pedestrian Dataset (PPD), a large-scale fisheye dataset for pedestrian detection in valet parking scenarios, addressing challenges like occlusions and diverse postures, and proposes data augmentation techniques to improve detection models.

Contribution

The paper presents the first large-scale fisheye pedestrian dataset for valet parking scenarios and introduces data augmentation methods to enhance detection performance.

Findings

Data augmentation improves detection accuracy.

Dataset demonstrates strong generalizability.

Baseline models benefit from the new dataset.

Abstract

Pedestrian detection under valet parking scenarios is fundamental for autonomous driving. However, the presence of pedestrians can be manifested in a variety of ways and postures under imperfect ambient conditions, which can adversely affect detection performance. Furthermore, models trained on publicdatasets that include pedestrians generally provide suboptimal outcomes for these valet parking scenarios. In this paper, wepresent the Parking Pedestrian Dataset (PPD), a large-scale fisheye dataset to support research dealing with real-world pedestrians, especially with occlusions and diverse postures. PPD consists of several distinctive types of pedestrians captured with fisheye cameras. Additionally, we present a pedestrian detection baseline on PPD dataset, and introduce two data augmentation techniques to improve the baseline by enhancing the diversity ofthe original dataset. Extensive experiments validate the effectiveness of our novel data augmentation approaches over baselinesand the dataset's exceptional generalizability.