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

DeepGlobe

Benchmarks

Task NameDataset NameSOTA ResultTrend
Few-Shot SegmentationDeepGlobe
mIoU62.5
83
Semantic SegmentationDeepglobe
mIoU89.5
35
Road SegmentationDeepGlobe
Precision89.55
24
Semantic SegmentationDeepGlobe (test)
mIoU74.1
20
Binary SegmentationDeepGlobe (test)
mIoU70.6
14
Cross-Domain Few-Shot SegmentationDeepGlobe (test)
mIoU63.4
12
Road SegmentationDeepGlobe (test)
Precision82.12
9
Semantic SegmentationDeepGlobeRoads
Dice Score75
7
Semantic SegmentationDeepGlobe (val)
mIoU64.38
7
Multi-label scene classificationDeepGlobe-ML Mixed Noise (test)
mAP macro (10% noise)79.15
7
Multi-label scene classificationDeepGlobe-ML Additive Noise (test)
mAP macro (10% noise)80.15
7
Few-Shot SegmentationDeepglobe (test)
mIoU53.1
3
Image SegmentationDeepglobe
mIoU44.31
2
Showing 13 of 13 rows