-
Notifications
You must be signed in to change notification settings - Fork 63
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Trajectory prediction results #16
Comments
Hi, have you find out the solution? I also want to use this project but I can not find a way to visualize the predicted trajectory. |
Hi, unfortunately, we do not have a way to do proper visualization. |
Do you know how I should use my own data set for trajectory prediction and evaluation, how can I find the data generated by the evaluation and use them to calculate relevant metrics? |
"Setup.py" resources download problem in TrackNPred |
|
Nah I haven't look at this project for a long time and I already forgot about that lol |
Have you found the new download link? Can you share it with me, thank you very much. |
First of all, congrats for the strong work here. However, as far as I inspected the code, the prediction module offers only an evaluation function, which doesn't offer much information. Is there an easy way to modify the code in order to see for each frame the center of the predicted objects, for example, in order to compare it with the actual center of the object? Also, the prediction is done for 3 seconds in the future? I want to make some experiments regarding trajectory prediction and it would be nice if I could use your project, but I need to vary the prediction time and also to see for each frame the predicted objects/ the predicted center for the objects, in order to compute the difference between the predicted center and the ground truth. I am interested specifically in car prediction. I see that you use Mean Suared Error and that the magic happens in maskedMSETest, however I don't really know what is the link between frames, object coordinates and the tensors x, y, muX, muY. Any help in order to detect the predicted objects and to change the prediction time will be greatly appreciated.
L.E.: How is the MSE computed? It takes only the detected objects and their centers?
The text was updated successfully, but these errors were encountered: