search

CVHub520 / X-AnyLabeling

Effortless data labeling with AI support from Segment Anything and other awesome models.
GNU General Public License v3.0
3.04k stars 345 forks source link

Auto-tracking Bounding Boxes Across Consecutive Frames #421

Open JansherMughal opened 1 month ago

JansherMughal commented 1 month ago

Hello,

I have a question regarding the functionality of x-anylabeling.

Is it possible to provide x-anylabeling with a bounding box (bbox) from one image and have it automatically apply this bbox to all subsequent images? Specifically, I am working with frames from a video, and I would like a bbox created on the first frame to be automatically built on the subsequent frames.

Thank you for your assistance.

Best regards, Jansher Mughal

CVHub520 commented 1 month ago

Hello @JansherMughal:

Thank you for reaching out and your interest in X-AnyLabeling.

We regret to inform you that the current version of X-AnyLabeling does not support the specific feature you mentioned, which is auto-tracking bounding boxes across consecutive frames. However, we do understand the importance and convenience of such functionality, especially for video object tracking applications.

While we cannot provide an immediate solution, we are open to integrating Video Object Segmentation (VOS) methods in future updates, which could potentially address your needs. We always prioritize the enhancement of our tools based on community feedback.

It also welcome contributions from the community. If you or anyone else is interested in adding this feature, I would be more than happy to review and merge a pull request.

Thank you for your understanding, and we look forward to a lively discussion on how we can improve X-AnyLabeling together.

Best regards, CVHub