It would be an excellent addition to have open-set object detectors such as Grounding Dino and/or YoloWorld which are capable of outputting bounding boxes on an image based on a user prompt. This would be amazing alongside MATLAB's implementation of SegmentAnything which already exists.
YoloWorld: https://docs.ultralytics.com/models/yolo-world/
Grounding Dino: https://github.com/IDEA-Research/GroundingDINO
It would be an excellent addition to have open-set object detectors such as Grounding Dino and/or YoloWorld which are capable of outputting bounding boxes on an image based on a user prompt. This would be amazing alongside MATLAB's implementation of SegmentAnything which already exists.
YoloWorld: https://docs.ultralytics.com/models/yolo-world/
Grounding Dino: https://github.com/IDEA-Research/GroundingDINO