Published on Wed Nov 25 2020

Temporal Autoencoder with U-Net Style Skip-Connections for Frame Prediction

Jay Santokhi, Pankaj Daga, Joned Sarwar, Anna Jordan, Emil Hewage

A traffic frame prediction approach that uses Convolutional LSTMs to create a Temporal Autoencoder. It uses U-Net style skip-connections that marry together recurrent and traditional computer vision techniques.

0
0
0
Abstract

Finding sustainable and novel solutions to predict city-wide mobility behaviour is an ever-growing problem given increased urban complexity and growing populations. This paper seeks to address this by describing a traffic frame prediction approach that uses Convolutional LSTMs to create a Temporal Autoencoder with U-Net style skip-connections that marry together recurrent and traditional computer vision techniques to capture spatio-temporal dependencies at different scales without losing topological details of a given city. Utilisation of Cyclical Learning Rates is also presented, improving training efficiency by achieving lower loss scores in fewer epochs than standard approaches.

Mon Oct 28 2019
Machine Learning
Recurrent Autoencoder with Skip Connections and Exogenous Variables for Traffic Forecasting
The increasing complexity of mobility plus the growing population in cities makes traffic forecasting an open and challenging task. In this paper, we introduce a novel approach to deal with predictions of speed, volume and main traffic direction. Exogenous variables like weather, time and calendar are also added in the model.
0
0
0
Mon Mar 01 2021
Artificial Intelligence
Listening to the city, attentively: A Spatio-Temporal Attention Boosted Autoencoder for the Short-Term Flow Prediction Problem
The importance of studying traffic flows and making predictions on alternative mobility (sharing services) has become increasingly important. We propose a framework, called STREED-Net, with multi-attention features to better mining the high-level spatial and temporal features.
0
0
0
Mon Jan 21 2019
Machine Learning
A deep learning approach to real-time parking occupancy prediction in spatio-temporal networks incorporating multiple spatio-temporal data sources
A deep learning model is applied for predicting block-level parking occupancy in real time. The model leverages Graph-Convolutional Neural Networks (GCNN) to extract spatial relations of traffic flow in large-scale networks. Weather data is particularly useful for improving predicting accuracy in recreational areas.
0
0
0
Mon Oct 12 2020
Machine Learning
Pedestrian Trajectory Prediction with Convolutional Neural Networks
Predicting the future trajectories of pedestrians is a challenging problem that has a range of application, from crowd surveillance to autonomous driving. In this work, we propose a new approach to pedestrian trajectory prediction, with the introduction of a novel 2Dconvolutional model.
0
0
0
Thu Jul 06 2017
Machine Learning
Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting
Diffusion Convolutional Recurrent Neural Network (DCRNN) is a deep learning framework for traffic forecasting. DCRNN incorporates both spatial and temporal dependency in the traffic flow. We evaluate the framework on two real-world large scale road network traffic datasets.
1
0
0
Thu Apr 16 2020
Machine Learning
Greenhouse Gas Emission Prediction on Road Network using Deep Sequence Learning
Predicting Greenhouse Gas (GHG) emissions is one of the profound topics, especially with the emergence of intelligent transportation systems (ITS) We develop a deep learning framework to predict link-level GHG emission rate (ER) (in CO2eq gram/second) based on the most representative predictors.
0
0
0