Yolov8s-segmentation not working on roboflow

Type: Yolov8s-segmentation
Link → Pixel Detector Object Detection Dataset and Pre-Trained Model by Pixel Data
OS → MacOs, Firefox.

I uploaded a model using the “uploading custom weights” documentation, and that worked fine. However, I can’t use it to annotate any image using the visualize tab. Yolov5(s) and Yolov8(s) detection (boxes) models both work, but not segmentation. I can confirm this works find on my own computer by running the .pt file. I want to be able to use it to assist with future annotations, and that requires a model uploaded to roboflow.

Hi @oliver408i - this is an object detection project type. Have you tried uploading the segmentation model to an instance segmentation project?

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.