PMMA: The Polytechnique Montreal Mobility Aids Dataset

2026-02-10

AI Summary

This study addresses the lack of fine-grained annotations for pedestrians using mobility aids—such as wheelchairs, canes, and walkers—in existing object detection datasets, which hinders accurate perception of vulnerable road users in intelligent transportation systems. To bridge this gap, we present the first systematically defined outdoor multi-class pedestrian detection dataset encompassing nine distinct pedestrian states, including wheelchair users, individuals pushing wheelchairs, and walker users. Leveraging the MMDetection framework, we evaluate seven state-of-the-art detection models—including YOLOX, Deformable DETR, and Faster R-CNN—and three tracking algorithms—ByteTrack, BOT-SORT, and OC-SORT—on this dataset. Experimental results indicate that YOLOX, Deformable DETR, and Faster R-CNN achieve the best detection performance, while the three trackers exhibit comparable accuracy. The dataset and code are publicly released to serve as a benchmark for future research.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Humans and AI: AI for Accessibility

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

This study introduces a new object detection dataset of pedestrians using mobility aids, named PMMA. The dataset was collected in an outdoor environment, where volunteers used wheelchairs, canes, and walkers, resulting in nine categories of pedestrians: pedestrians, cane users, two types of walker users, whether walking or resting, five types of wheelchair users, including wheelchair users, people pushing empty wheelchairs, and three types of users pushing occupied wheelchairs, including the entire pushing group, the pusher and the person seated on the wheelchair. To establish a benchmark, seven object detection models (Faster R-CNN, CenterNet, YOLOX, DETR, Deformable DETR, DINO, and RT-DETR) and three tracking algorithms (ByteTrack, BOT-SORT, and OC-SORT) were implemented under the MMDetection framework. Experimental results show that YOLOX, Deformable DETR, and Faster R-CNN achieve the best detection performance, while the differences among the three trackers are relatively small. The PMMA dataset is publicly available at https://doi.org/10.5683/SP3/XJPQUG, and the video processing and model training code is available at https://github.com/DatasetPMMA/PMMA.

Problem

• mobility aids
• pedestrian detection
• object detection dataset
• wheelchair users
• assistive devices

Innovation

• mobility aids dataset
• fine-grained pedestrian detection
• outdoor object detection
• wheelchair user recognition
• benchmark evaluation