A Universal Railway Obstacle Detection System based on Semi-supervised Segmentation And Optical Flow

TL;DR

This paper introduces a semi-supervised segmentation method guided by optical flow and synthetic data generation to detect railway obstacles across diverse conditions, addressing out-of-distribution challenges effectively.

Contribution

It proposes a novel binary segmentation framework using optical flow and synthetic data, improving obstacle detection without extensive manual annotations.

Findings

Effective obstacle detection across varying conditions

Synthetic data reduces manual annotation needs

Optical flow enhances segmentation accuracy

Abstract

Detecting obstacles in railway scenarios is both crucial and challenging due to the wide range of obstacle categories and varying ambient conditions such as weather and light. Given the impossibility of encompassing all obstacle categories during the training stage, we address this out-of-distribution (OOD) issue with a semi-supervised segmentation approach guided by optical flow clues. We reformulate the task as a binary segmentation problem instead of the traditional object detection approach. To mitigate data shortages, we generate highly realistic synthetic images using Segment Anything (SAM) and YOLO, eliminating the need for manual annotation to produce abundant pixel-level annotations. Additionally, we leverage optical flow as prior knowledge to train the model effectively. Several experiments are conducted, demonstrating the feasibility and effectiveness of our approach.