Masks instead of bounding boxes #6747
-
Hi, I'm new to YOLOv5 and object detection in general. Is there an option for YOLOv5 to output masks instead of bounding boxes for classification? If not, can you advise on what is the proper workflow following training using YOLOv5 to get those masks? Specifically, I would like to get the surface area of the object detected. Thanks! |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments
-
@thepaulchee hi thanks for asking about masks! Segmentation (masks) is in our product pipeline for later this year. Currently YOLOv5 supports classification (with classification branch) and detection. |
Beta Was this translation helpful? Give feedback.
-
Thanks! Good to know it's in the pipeline! 👍🏻 |
Beta Was this translation helpful? Give feedback.
-
Is this pipeline done ? |
Beta Was this translation helpful? Give feedback.
@thepaulchee hi thanks for asking about masks! Segmentation (masks) is in our product pipeline for later this year. Currently YOLOv5 supports classification (with classification branch) and detection.