SAMRS: Scaling-up Remote Sensing Segmentation Dataset with Segment Anything Model: Supplementary Material

A.1 Category Abbreviations For the SOTA dataset, we present the list of all category abbreviations as follows. For the SIOR dataset, we present the list of all category abbreviations as follows. For the FAST dataset, we present the list of all category abbreviations as follows. B.1 Experiment Settings We present the experiment settings of pre-training and fine-tuning in Table S1-S2. This work was partially done during Di Wang's internship at iFlytek. "|" means Table S1: Basic settings in experiments. B.2 SAMRS Training and Validation sets For the experiments based on the SAMRS dataset (see Table 2 in the main text), in each subset, 95% samples are used for pre-training.