TPE-Net: Track Point Extraction and Association Network for Rail Path Proposal Generation

TL;DR

TPE-Net is a convolutional neural network designed to extract and associate rail path pixels from images, enabling autonomous trains to identify potential routes and obstacles without relying on 3D data or prior geometric knowledge.

Contribution

The paper introduces TPE-Net, a novel fully convolutional encoder-decoder architecture for rail path extraction and association, improving route hypothesis generation in train control systems.

Findings

Achieves 0.9207 precision and 0.8721 recall in rail pixel extraction.

Operates at approximately 12 frames per second.

Performs well without dependence on camera parameters or 3D data.

Abstract

One essential feature of an autonomous train is minimizing collision risks with third-party objects. To estimate the risk, the control system must identify topological information of all the rail routes ahead on which the train can possibly move, especially within merging or diverging rails. This way, the train can figure out the status of potential obstacles with respect to its route and hence, make a timely decision. Numerous studies have successfully extracted all rail tracks as a whole within forward-looking images without considering element instances. Still, some image-based methods have employed hard-coded prior knowledge of railway geometry on 3D data to associate left-right rails and generate rail route instances. However, we propose a rail path extraction pipeline in which left-right rail pixels of each rail route instance are extracted and associated through a fully convolutional encoder-decoder architecture called TPE-Net. Two different regression branches for TPE-Net are proposed to regress the locations of center points of each rail route, along with their corresponding left-right pixels. Extracted rail pixels are then spatially clustered to generate topological information of all the possible train routes (ego-paths), discarding non-ego-path ones. Experimental results on a challenging, publicly released benchmark show true-positive-pixel level average precision and recall of 0.9207 and 0.8721, respectively, at about 12 frames per second. Even though our evaluation results are not higher than the SOTA, the proposed regression pipeline performs remarkably in extracting the correspondences by looking once at the image. It generates strong rail route hypotheses without reliance on camera parameters, 3D data, and geometrical constraints.