WISE-Net

Reported on 6 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision6 results

Instance SegmentationonCOCO 2017 val
AP· uses extra data
7.8
best: 45.1 (SparK (ConvNeXt V1-B Mask R-CNN))
Instance SegmentationonCOCO 2017 val
AP@50· uses extra data
18.2
best: 38.4 (WeakSAM-Mask2Former (with SAM))
Instance SegmentationonCOCO 2017 val
AP@75· uses extra data
8.8
best: 27 (WeakSAM-Mask2Former (with SAM))
Instance SegmentationonPASCAL VOC 2012 val
mAP@0.25· uses extra data
53.5
best: 73.4 (WeakSAM-Mask2Former (with SAM))
Instance SegmentationonPASCAL VOC 2012 val
mAP@0.5· uses extra data
43
best: 64.4 (WeakSAM-Mask2Former (with SAM))
Instance SegmentationonPASCAL VOC 2012 val
mAP@0.75· uses extra data
25.9
best: 45.3 (WeakSAM-Mask2Former (with SAM))