Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

RRSIS-D

Benchmarks

Task NameDataset NameSOTA ResultTrend
Referring Remote Sensing Image SegmentationRRSIS-D (test)
gIoU69.17
33
Referring Expression Comprehension (RSREC)RRSIS-D (test)
Pr@0.576.96
29
Referring Expression SegmentationRRSIS-D
mIoU71.08
27
Referring Remote Sensing Image SegmentationRRSIS-D (val)
mIoU (Mean IoU)68.1
23
Region-level ClassificationRRSIS-D
Top-1 Accuracy82.89
21
Remote Sensing Visual GroundingRRSIS-D official (test)
Acc@0.565.73
14
Visual GroundingRRSIS-D
mIoU76.66
10
Referring Expression SegmentationRRSIS-D (test)
gIoU50.97
8
Remote Sensing Visual GroundingRRSIS-D (SOTA-T)
Accuracy @0.532.79
5
Showing 9 of 9 rows