Stan

Interested too, my DICOMs are 2D videos by default - currently I'm converting them to a video extension but would be interested to see this implemented natively.

Any updates on this? There have been a lot of performance improvements in recent implementations of SAM such as [EfficientSAM](https://github.com/mit-han-lab/efficientvit) reducing the hardware requirements to run models for pre labelling.

Also interested in a feature like this. The open source [VGG Image Annotator](https://github.com/ox-vgg/via/tree/master) has something that could potentially be used as a starting point. https://www.robots.ox.ac.uk/~vgg/software/via/demo/via_video_annotator.html