Investigate SAM 3 for combined detection and segmentation; integrate if viable

SAM 3 may provide greater morphological and general specimen layout and species flexibility with text-based segmentation.

Resource requirements will certainly be greater, but if performance excels on a greater variety of inputs, it would be a valuable addition to complement instructions on producing custom YOLO models.

Start by reviewing the project's current custom YOLO-model instructions and evaluating SAM 3 against the stated specimen-layout, species-flexibility, and resource requirements. Done means documenting whether SAM 3 is viable and, if so, integrating it as a complementary detection and segmentation option.

Written by the indexing model from the issue text.