Skip to content
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

About Top-View Coordinates. #34

Open
changjo opened this issue Sep 23, 2022 · 7 comments
Open

About Top-View Coordinates. #34

changjo opened this issue Sep 23, 2022 · 7 comments
Assignees
Labels
bug Something isn't working question Further information is requested

Comments

@changjo
Copy link

changjo commented Sep 23, 2022

Hi Atom,
I have downloaded the datasets and tried to visualize them using the visualize_frame function but the top-view coordinates don't seem to match the video.
And I have observed the drone sometimes rotates itself a little bit in a video but it seems that the dron_keypoints.json doesn't consider it.
I have wanted to convert top-view coordinates to local pitch coordinates and use them as ground truths.
I was wondering if I have missed some steps.

tv_output

@changjo changjo changed the title About Top-View Coordinates Issue. About Top-View Coordinates. Sep 23, 2022
@AtomScott AtomScott self-assigned this Sep 23, 2022
@AtomScott
Copy link
Owner

Thanks for the question!
The steps you took are correct and should work so it seems like I may have made a mistake in uploading the data (it looks as though the bounding boxes are from the wide-view camera).

I can't re-upload immediately, but I'll get to work after the weekend.

@changjo
Copy link
Author

changjo commented Sep 23, 2022

I see. Thanks a lot.

@AtomScott AtomScott added question Further information is requested bug Something isn't working labels Sep 25, 2022
@AtomScott
Copy link
Owner

Okay, so I looked into this and there seem to be two problems.

  1. I uploaded the wide-view annotations as top-view camera annotations by mistake
  2. The top-view video has a different frame rate than the wide-view video, but it seems as though I may have preprocessed the annotations using the same frame rate.

I have solved problem 1 and am working on problem 2, so please be patient for a little longer.
Also, @IkumaUchida is implementing API for visualizing bboxes on video so we can confirm that the annotations are correct visually 👀

@changjo
Copy link
Author

changjo commented Sep 27, 2022

Thanks a lot!

@mrba59
Copy link

mrba59 commented Nov 5, 2022

Hi,
First of all thank you for this amazing dataset , i was looking for this type of data since a while.
Did you fixed this problem ? I still have the same issue when i 'am trying to draw bounding box of the top_view

@AtomScott
Copy link
Owner

Hi, sorry for the delay!

After receving reports of erroneous data, we have fixed and reuploaded a majority of SoccerTrack. We are also adding videos with visualized bounding boxes so that you can be sure that the data is good. The visualizations can be found in the viz_results directory under Top-view/Wide-view (see Kaggle).

However, there is still work to do. In the meantime, we have created a spreadsheet to keep everyone updated on our progress: Spreadsheet Link

Can you try download the data again and see if the problems persists?

@mrba59
Copy link

mrba59 commented Nov 12, 2022

Hi, sorry for the delay!

After receving reports of erroneous data, we have fixed and reuploaded a majority of SoccerTrack. We are also adding videos with visualized bounding boxes so that you can be sure that the data is good. The visualizations can be found in the viz_results directory under Top-view/Wide-view (see Kaggle).

However, there is still work to do. In the meantime, we have created a spreadsheet to keep everyone updated on our progress: Spreadsheet Link

Can you try download the data again and see if the problems persists?

Everything looks good with the new data, thanks a lot !!
I have another question , do you have the four corner point positions's of the field or any other marker like center circle or the penalty marker or the real dimension of the field ? I noticed sometime that the drone move a little bit so the KPI like speed trajectory will be affected

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working question Further information is requested
Projects
None yet
Development

No branches or pull requests

3 participants