CyclingNet is a deep computer vision model based on convolutionalstructure embedded with self-attention bidirectional long-short term memory blocks . It aims to understand near misses from both sequential images of scenes and their optical flows . The model is trained on scenes of both saferides and near misses . After 42 hours of training on a single GPU, the modelshows high accuracy on the training, testing and validation sets. The model can be pipelined with other state-of-the-art classifiers and objectdetectors simultaneously to understand the causality of near misses based on interactions of road-users, the built and the naturalenvironments . It is intended to be used for generating information that can draw significant conclusions regarding cycling behaviour in cities and elsewhere, which could help planners and policy-makers to better understand the safety of cycling behaviour, which can be used to help planners, analysts and engineers alike . Back to Mail Online Online home .Back to the page you came from: http://www.mailonline.com/news/science-back-to-date .

Author(s) : Mohamed R. Ibrahim, James Haworth, Nicola Christie, Tao Cheng

Links : PDF - Abstract

Code :
Coursera

Keywords : misses - scenes - model - understand - cycling -

Leave a Reply

Your email address will not be published. Required fields are marked *