I have a dataset with 4 channel input imagery, 2 labels set - one label set is in MS COCO JSON Format and another labels format is in TIF format.
The dataset is designed for instance segmentation multi-modal multi-task.
How do I give input of 2 labels for maskrcnn model?
How do I tell the model to do multi-tasking?
Nandtiw
commented
1 year ago
I have a dataset with 4 channel input imagery, 2 labels set - one label set is in MS COCO JSON Format and another labels format is in TIF format. The dataset is designed for instance segmentation multi-modal multi-task. How do I give input of 2 labels for maskrcnn model? How do I tell the model to do multi-tasking?
Can someone please provide guidance on this task?