Add mask for instance segmentation

[arashazimi0032]

Hello, I saw in your code that in the yolo v8 model, on line 62, you save only the output boxes from the model, it would be great if you also save the masks and display them in the output.

sahi/sahi/models/yolov8.py

Line 62 in 0414e47

prediction_result = [

prediction_result = [
result.boxes.data[result.boxes.data[:, 4] >= self.confidence_threshold] for result in prediction_result
]

[fcakyon]

Segmentation mask support for yolov8, mmdetection, detectron2, torchvision models is live with the latest release! @FilipeMendesVMD: https://github.com/obss/sahi/discussions/1051

There is also 5x memory and speed improvement when working with masks!

Check the demo notebook on how it works: https://github.com/obss/sahi/blob/main/demo/inference_for_yolov8.ipynb