NSegment : Label-specific Deformations for Remote Sensing Image Segmentation
This addresses annotation inconsistencies in remote sensing for researchers and practitioners, but it is incremental as it builds on existing augmentation techniques.
The paper tackles labeling errors and data scarcity in remote sensing image segmentation by proposing NSegment, a data augmentation method that applies label-specific elastic deformations, which improves performance over state-of-the-art models.
Labeling errors in remote sensing (RS) image segmentation datasets often remain implicit and subtle due to ambiguous class boundaries, mixed pixels, shadows, complex terrain features, and subjective annotator bias. Furthermore, the scarcity of annotated RS data due to the high cost of labeling complicates training noise-robust models. While sophisticated mechanisms such as label selection or noise correction might address the issue mentioned above, they tend to increase training time and add implementation complexity. In this paper, we propose NSegment-a simple yet effective data augmentation solution to mitigate this issue. Unlike traditional methods, it applies elastic transformations only to segmentation labels, varying deformation intensity per sample in each training epoch to address annotation inconsistencies. Experimental results demonstrate that our approach improves the performance of RS image segmentation over various state-of-the-art models.