-
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
Add track_id
to CVAT for Images, correct importing tracks from CVAT for Images
#7239
Comments
Please, use CVAT for Images. |
But i need track id also. As far as I saw, CVAT for images do not support Track ID feature |
There were request to add |
For now you may try Datumaro format. |
track_id
to CVAT for Images, correct importing tracks from CVAT for Images
Can you otherwise tell me in what order are the frames numbered, if we export it using the CVAT for Videos format. Is it sorted based on filename? or can we give a sort function for it? |
Datumaro also has the same problem as importing COCO with track attributes. Each box is considered as a new track, even though there is a track attribute in annotation section. |
Just tried, it restored tracks on current develop branch. |
Oh I forgot to mention I was using CVAT 2.3 version. Which format worked Datumaro or COCO? Also, does the current master branch have the same feature? |
@bsekachev |
Probably, yes. Anyway we will make new release these days. |
That would be helpful. Thanks |
Hi, This wasn't supposed to work with this format, as there is the CVAT for video, designed to support tracks specifically. However, as we added this in other image-based formats, such as VOC as COCO, and supported track import in those cases, we can align behavior here as well, and include track_id attributes in CVAT for images, when tracks are exported. |
Inclusion of image name/path would be helpful because, we dont know how the images are stored in ascending or descending order in cvat folder. We just receive frame numbers from export, which is unclear and non-descriptive.
Thanks
Originally posted by @jeethesh-pai in #4502 (comment)
The text was updated successfully, but these errors were encountered: