How should I handle label(void label) to get accurate crop semantic segmentation results?For semantic segmentation, I think category 19 is a mixture of different crops.In the article "Multi-modal temporal attention models for crop mapping from satellite time series," you mentioned that "In this setting,the models also predict background pixels, resulting in a K = 19
How can I achieve the same result?
Looking forward to your reply
How should I handle label(void label) to get accurate crop semantic segmentation results?For semantic segmentation, I think category 19 is a mixture of different crops.In the article "Multi-modal temporal attention models for crop mapping from satellite time series," you mentioned that "In this setting,the models also predict background pixels, resulting in a K = 19 How can I achieve the same result? Looking forward to your reply