search

z-x-yang / Segment-and-Track-Anything

An open-source project dedicated to tracking and segmenting any objects in videos, either automatically or interactively. The primary algorithms utilized include the Segment Anything Model (SAM) for key-frame segmentation and Associating Objects with Transformers (AOT) for efficient tracking and propagation purposes.
GNU Affero General Public License v3.0
2.84k stars 341 forks source link

Segmenting objects based on given bounding box #38

Closed kulkarnikeerti closed 1 year ago

kulkarnikeerti commented 1 year ago

Hi @yamy-cheng , @yoxu515 , @z-x-yang , @lingorX

Thanks again for the great work. I know that we can give input in text format or stroke or even click on the object which we would like to track. I would like to know if its possible with the current code base to use the object's bounding box as an input to track those objects?

Really appreciate your response. Thanks

yamy-cheng commented 1 year ago

Do you mean user input a coordinate of box to select object?

kulkarnikeerti commented 1 year ago

@yamy-cheng , yes.

yamy-cheng commented 1 year ago

@yamy-cheng , yes.

The current WebUI does not support this feature, but we will add a notebook for this in the future.

kulkarnikeerti commented 1 year ago

@yamy-cheng Okay. Thanks. Waiting for the notebook:)