Research Article
Predicting Fine-Grained Traffic Conditions via Spatio-Temporal LSTM
Table 4
Applications and comparisons about the traffic prediction.
| Year | Literature | Basic data source | Target | Term | Core algorithm | Complexity | Granularity | Object-based |
| 2018 | [18] | web camera | traffic density | short | Convolutional neural network | high | fine-grained | intersection | [27] | an open dataset | traffic flow | long/short | Generative adversarial network | high | coarse-grained | freeway |
| 2017 | [17] | web camera | traffic density | short | Fully convolutional networks | high | fine-grained | restricted area | [15] | an experimental car | vehicle speed | short | Auto-regressive model | middle | fine-grained | road segment | [14] | floating car | vehicle speed | short | HMMs+SUMO | middle | coarse-grained | motorway | [13] | Loop Detector | traffic volume | short | ST semi-supervised learning | low | fine-grained | road segment | [1] | traffic loops | traffic flow | long | Gaussian process regression | low | coarse-grained | region |
|
|