Spatial-Temporal Attentive LSTM for Vehicle-Trajectory Prediction
Round 1
Reviewer 1 Report
In this paper, authors have proposed a novel deep learning-based model for vehicle trajectory prediction. The proposed model has captured the spatial-temporal interactions and motion feature of the target vehicle. The proposed model is validated on the public datasets and experiements demonstrate better performance. The paper in current version can be accepted.
Author Response
Dear reviewers,
Thank you very much for your comments and approvals.
Best regards,
Rui Jiang,
June 16, 2022
Reviewer 2 Report
The paper entitled "Spatial-Temporal Attentive LSTM for Vehicle Trajectory Prediction" proposes a spatial-temporal attentive LSTMencoder-decoder model (STAM-LSTM) to predict vehicle trajectories.
In summary, the paper reads well and is well organized. The essential components (argument for a new method, contributions, related works, etc) can be easily found through the text.
However, it is essential to address the following:
1. In the related works, what I feel is missing in the three subsections is a conclusion regarding the findings. Essentially, how do you place your work among the ones discussed? What does your work do that is not addressed by the other works?
2. In the experiments, it is important to inform the standard deviation (Tables 1 and 2).
3. In section 4.7, the discussion of the qualitative analysis is very raw. I recommend the authors to go more in-depth and use more examples.
Author Response
Dear reviewers,
Thank you very much for your questions. Our detailed modifications are listed in the attachments.
Best regards,
Rui Jiang,
June 16, 2022
Author Response File: Author Response.pdf
Reviewer 3 Report
Authors should better describe the usefulness of predicting vehicle trajectories.
Author Response
Dear reviewers,
Thank you very much for your questions. Our detailed modifications are listed in the attachments.
Best regards,
Rui Jiang,
June 16, 2022
Author Response File: Author Response.pdf