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

Too many frames have rectangles when using a Tracker #3796

Closed
AmitN1212 opened this issue Oct 13, 2021 · 13 comments
Closed

Too many frames have rectangles when using a Tracker #3796

AmitN1212 opened this issue Oct 13, 2021 · 13 comments
Labels
need info Need more information to investigate the issue

Comments

@AmitN1212
Copy link
Contributor

Hi, I am trying to use the built in tracker model (pytorch foolwood siamask). It works well but the problem is that even when I set the tracking frame number, it keeps putting rectangles inside the remaining frames. Is there any way to solve it?
Thank you.

@bsekachev
Copy link
Member

Hi @AmitN1212

Tracking was completely redesigned. Please, update your version.

@AmitN1212
Copy link
Contributor Author

Thank you. I just installed the last version of CVAT. However, the tracker is not working right now, I get an error about an internal error. Do you know what is the problem?

@bsekachev
Copy link
Member

Probably you need to redeploy serverless function, because its API has been changed

@AmitN1212
Copy link
Contributor Author

I redployed it. I am still getting the same error. I am using the built in foolwood model, with the instructions that are here

@azhavoro
Copy link
Contributor

@AmitN1212 Try to delete the function with nuctl delete function <function_name> firstly.

@AmitN1212
Copy link
Contributor Author

Still not working. I tried to debug the code and it seems that in this block:

    shapes = data.get("shapes")
    states = data.get("states")
    image = Image.open(buf)

    results = {
        'shapes': [],
        'states': []
    }
    for i, shape in enumerate(shapes):

Shapes is None so it does not work. If it is fine, what else can I try?

@bsekachev
Copy link
Member

If shapes is none, it probably means that your function is not updated. Try to check output from nuclio-dashboard container.

@AmitN1212
Copy link
Contributor Author

What do you mean? I took the function from CVAT repo. How can I see the output of the container?

@bsekachev
Copy link
Member

docker logs nuclio-dashboard

@bsekachev
Copy link
Member

Any updates?

@bsekachev bsekachev added the need info Need more information to investigate the issue label Oct 26, 2021
@AmitN1212
Copy link
Contributor Author

It says that I don't have a container called nuclio-dashboard

@bsekachev
Copy link
Member

@AmitN1212

Just run it:

docker run -p 8070:8070 -v /var/run/docker.sock:/var/run/docker.sock -v /tmp:/tmp --name nuclio-dashboard quay.io/nuclio/dashboard:stable-amd64

@AmitN1212
Copy link
Contributor Author

Thank you. I probably had some problems with my photos type. However, I had problems with all the images types in the release version. Currently, I am using the development environment version.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
need info Need more information to investigate the issue
Projects
None yet
Development

No branches or pull requests

3 participants