-
Notifications
You must be signed in to change notification settings - Fork 3.1k
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
How to upload a task from a video with annotations? #4325
Comments
Hi, What you need is probably backup/restore of a task. You can disable "Reset zoom" (enabled by default for images): There is related bug #4289, but generally it works |
Thank you for response! Can you share a latest version docker images with the backup feature? I'm not able to build them myself so far on my machine |
Why can't you build it? Usually we publish docker images when releases come. |
@bsekachev its just tricky to set everything up on Mac with Apple processor in order to build. |
I am not sure I understand what kind of re-import you are talking about. Could you please clarify the scenario? If it is true, it might be a bug, because I don't expect image quality degradation after any kind of import. |
@bsekachev the scenario is following:
In the imported task quality is significantly worse |
@evekeen Are you able to provide a couple of screenshots with these images? If image quality is actually worse, better to submit a new issue with the scenario you provided and attach these screenshot, because as I told before, I suspect it is just a bug |
@bsekachev I've created an issue: #4377 |
Use case:
There is a way of importing annotations along with a video through CLI/REST, although the annotations have frame_XXX references, and the upload only works with a set of images.
Is there a way of creating a task via CLI using a video file and the dumped annotations?
So far I managed to split video into images first and then upload as image sequence. But then zoom always resets in the UI, since it interprets the task as per-image, which actually is not true. And I cannot guarantee that my image sequence is the same as it was at the other person instance.
The text was updated successfully, but these errors were encountered: