paper-with-me

홈 › Papers

RailSafeNet: Visual Scene Understanding for Tram Safety

2025-09-15 · Ondřej Valach, Ivan Gruber arxiv

Tram-human interaction safety is an important challenge, given that trams frequently operate in densely populated areas, where collisions can range from minor injuries to fatal outcomes. This paper addresses the issue from the perspective of designing a solution leveraging digital image processing, deep learning, and artificial intelligence to improve the safety of pedestrians, drivers, cyclists, pets, and tram passengers. We present RailSafeNet, a real-time framework that fuses semantic segmentation, object detection and a rule-based Distance Assessor to highlight track intrusions. Using only monocular video, the system identifies rails, localises nearby objects and classifies their risk by comparing projected distances with the standard 1435mm rail gauge. Experiments on the diverse RailSem19 dataset show that a class-filtered SegFormer B3 model achieves 65% intersection-over-union (IoU), while a fine-tuned YOLOv8 attains 75.6% mean average precision (mAP) calculated at an intersection over union (IoU) threshold of 0.50. RailSafeNet therefore delivers accurate, annotation-light scene understanding that can warn drivers before dangerous situations escalate. Code available at https://github.com/oValach/RailSafeNet.

📄 PDF Abstract BibTeX arXiv:2509.12125

Code (0)

등록된 구현이 없습니다.

Tasks

Semantic SegmentationScene UnderstandingObject Detection

Similar Papers 제목 키워드 기반

RailSem19: A Dataset for Semantic Rail Scene Understanding

2019-06-16 · CVPR 2019 6 · Oliver Zendel, Markus Murschitz, Marcel Zeilinger, Daniel Steininger 외

Solving tasks for autonomous road vehicles using com-puter vision is a dynamic and active research field. How-ever, one aspect of autonomous transportation has receivedlittle contributions: the rail domain. In this paper…

Scene UnderstandingSemantic SegmentationTransfer Learning

TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos

2024-03-26 · Yufu Wang, ZiYun Wang, Lingjie Liu, Kostas Daniilidis

We propose TRAM, a two-stage method to reconstruct a human's global trajectory and motion from in-the-wild videos. TRAM robustifies SLAM to recover the camera motion in the presence of dynamic humans and uses the scene b…

3D Human Pose Estimation

Seeing Without Eyes: 4D Human-Scene Understanding from Wearable IMUs

2026-04-23 · Hao-Yu Hsu, Tianhang Cheng, Jing Wen, Alexander G. Schwing 외 arxiv

Understanding human activities and their surrounding environments typically relies on visual perception, yet cameras pose persistent challenges in privacy, safety, energy efficiency, and scalability. We explore an altern…

Scene Understanding

eTraM: Event-based Traffic Monitoring Dataset

2024-03-29 · CVPR 2024 1 · Aayush Atul Verma, Bharatesh Chakravarthi, Arpitsinh Vaghela, Hua Wei 외

Event cameras, with their high temporal and dynamic range and minimal memory usage, have found applications in various fields. However, their potential in static traffic monitoring remains largely unexplored. To facilita…

PreTraM: Self-Supervised Pre-training via Connecting Trajectory and Map

2022-04-21 · Chenfeng Xu, Tian Li, Chen Tang, Lingfeng Sun 외

Deep learning has recently achieved significant progress in trajectory forecasting. However, the scarcity of trajectory data inhibits the data-hungry deep-learning models from learning good representations. While mature …

Contrastive LearningRepresentation LearningTrajectory Forecasting