Classifying Bicycle Infrastructure Using On-Bike Street-Level Images

• URL: http://arxiv.org/abs/2410.19194v1
• Date: Thu, 24 Oct 2024 22:58:31 GMT
• Title: Classifying Bicycle Infrastructure Using On-Bike Street-Level Images
• Authors: Kal Backman, Ben Beck, Dana Kulić,
• Abstract summary: Many potential cyclists are discouraged from taking up cycling due to the lack of suitable and safe infrastructure. We propose a system capable of classifying available cycling infrastructure from on-bike smartphone camera data. This work is the first to classify cycling infrastructure using only street-level imagery collected from bike-mounted mobile phone cameras.
• Score: 0.0
• License:
• Abstract: While cycling offers an attractive option for sustainable transportation, many potential cyclists are discouraged from taking up cycling due to the lack of suitable and safe infrastructure. Efficiently mapping cycling infrastructure across entire cities is necessary to advance our understanding of how to provide connected networks of high-quality infrastructure. Therefore we propose a system capable of classifying available cycling infrastructure from on-bike smartphone camera data. The system receives an image sequence as input, temporally analyzing the sequence to account for sparsity of signage. The model outputs cycling infrastructure class labels defined by a hierarchical classification system. Data is collected via participant cyclists covering 7,006Km across the Greater Melbourne region that is automatically labeled via a GPS and OpenStreetMap database matching algorithm. The proposed model achieved an accuracy of 95.38%, an increase in performance of 7.55% compared to the non-temporal model. The model demonstrated robustness to extreme absence of image features where the model lost only 6.6% in accuracy after 90% of images being replaced with blank images. This work is the first to classify cycling infrastructure using only street-level imagery collected from bike-mounted mobile phone cameras, while demonstrating robustness to feature sparsity via long temporal sequence analysis.

Related papers

• Evaluating the effects of Data Sparsity on the Link-level Bicycling Volume Estimation: A Graph Convolutional Neural Network Approach [54.84957282120537]
  We present the first study to utilize a Graph Convolutional Network architecture to model link-level bicycling volumes. We estimate the Annual Average Daily Bicycle (AADB) counts across the City of Melbourne, Australia using Strava Metro bicycling count data. Our results show that the GCN model performs better than these traditional models in predicting AADB counts.
  arXiv  Detail & Related papers  (2024-10-11T04:53:18Z)
• Homography Guided Temporal Fusion for Road Line and Marking Segmentation [73.47092021519245]
  Road lines and markings are frequently occluded in the presence of moving vehicles, shadow, and glare. We propose a Homography Guided Fusion (HomoFusion) module to exploit temporally-adjacent video frames for complementary cues. We show that exploiting available camera intrinsic data and ground plane assumption for cross-frame correspondence can lead to a light-weight network with significantly improved performances in speed and accuracy.
  arXiv  Detail & Related papers  (2024-04-11T10:26:40Z)
• Bent & Broken Bicycles: Leveraging synthetic data for damaged object re-identification [59.80753896200009]
  We propose a novel task of damaged object re-identification, which aims at distinguishing changes in visual appearance due to deformations or missing parts from subtle intra-class variations. We leverage the power of computer-generated imagery to create, in a semi-automatic fashion, high-quality synthetic images of the same bike before and after a damage occurs. As a baseline for this task, we propose TransReI3D, a multi-task, transformer-based deep network unifying damage detection.
  arXiv  Detail & Related papers  (2023-04-16T20:23:58Z)
• BikeDNA: A Tool for Bicycle Infrastructure Data & Network Assessment [0.0]
  BikeDNA is an open-source tool for reproducible quality assessment of bicycle infrastructure data. BikeDNA supports quality assessments of bicycle infrastructure data for a wide range of applications.
  arXiv  Detail & Related papers  (2023-03-02T13:06:59Z)
• Monocular Cyclist Detection with Convolutional Neural Networks [0.0]
  This study aims to reduce the number of vehicle-cyclist collisions, which are often caused by poor driver attention to blind spots. We designed a state-of-the-art real-time monocular cyclist detection that can detect cyclists with object detection convolutional neural networks. We conclude that this cyclist detection device can accurately and quickly detect cyclists and has the potential to improve cyclist safety significantly.
  arXiv  Detail & Related papers  (2023-01-16T13:54:13Z)
• Predicting Citi Bike Demand Evolution Using Dynamic Graphs [81.12174591442479]
  We apply a graph neural network model to predict bike demand in the New York City, Citi Bike dataset. In this paper, we attempt to apply a graph neural network model to predict bike demand in the New York City, Citi Bike dataset.
  arXiv  Detail & Related papers  (2022-12-18T21:43:27Z)
• Mapping suburban bicycle lanes using street scene images and deep learning [0.0]
  This thesis presents a method to create a map of bicycle lanes in a survey area by taking sample street scene images from each road. A deep learning model that has been trained to recognise bicycle lane symbols is applied. The method was applied to successfully build a map for a survey area in the outer suburbs of Melbourne.
  arXiv  Detail & Related papers  (2022-04-27T04:56:26Z)
• Automated Detection of Missing Links in Bicycle Networks [0.15293427903448023]
  We develop the IPDC procedure (Identify, Prioritize, Decluster, Classify) for finding the most important missing links in urban bicycle networks. We first identify all possible gaps following a multiplex network approach, prioritize them according to a flow-based metric, decluster emerging gap clusters, and manually classify the types of gaps. Our results show how network analysis with minimal data requirements can serve as a cost-efficient support tool for bicycle network planning.
  arXiv  Detail & Related papers  (2022-01-10T15:35:14Z)
• CyclingNet: Detecting cycling near misses from video streams in complex urban scenes with deep learning [1.462434043267217]
  CyclingNet is a deep computer vision model based on convolutional structure embedded with self-attention bidirectional long-short term memory (LSTM) blocks. After 42 hours of training on a single GPU, the model shows high accuracy on the training, testing and validation sets. The model is intended to be used for generating information that can draw significant conclusions regarding cycling behaviour in cities.
  arXiv  Detail & Related papers  (2021-01-31T23:59:28Z)
• Dynamic Bicycle Dispatching of Dockless Public Bicycle-sharing Systems using Multi-objective Reinforcement Learning [79.61517670541863]
  How to use AI to provide efficient bicycle dispatching solutions based on dynamic bicycle rental demand is an essential issue for dockless PBS (DL-PBS) We propose a dynamic bicycle dispatching algorithm based on multi-objective reinforcement learning (MORL-BD) to provide the optimal bicycle dispatching solution for DL-PBS.
  arXiv  Detail & Related papers  (2021-01-19T03:09:51Z)
• Dynamic Planning of Bicycle Stations in Dockless Public Bicycle-sharing System Using Gated Graph Neural Network [79.61517670541863]
  Dockless Public Bicycle-sharing (DL-PBS) network becomes increasingly popular in many countries. redundant and low-utility stations waste public urban space and maintenance costs of DL-PBS vendors. We propose a Bicycle Station Dynamic Planning (BSDP) system to dynamically provide the optimal bicycle station layout for the DL-PBS network.
  arXiv  Detail & Related papers  (2021-01-19T02:51:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.