Mask R-CNN (ResNet-50-FPN)

Reported on 13 benchmarks across 8 tasks · 2 papers · 2 SOTA

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

Methodology8 results

3DonCOCO minival
box AP· 2017-03-20
37.7
best: 66 (PE_spatial (DETA))
Mask R-CNN arXiv:1703.06870
2D ClassificationonCOCO minival
box AP· 2017-03-20
37.7
best: 66 (PE_spatial (DETA))
Mask R-CNN arXiv:1703.06870
2D Object DetectiononCOCO minival
box AP· 2017-03-20
37.7
best: 66 (PE_spatial (DETA))
Mask R-CNN arXiv:1703.06870
16konCOCO minival
box AP· 2017-03-20
37.7
best: 66 (PE_spatial (DETA))
Mask R-CNN arXiv:1703.06870
3DonTBBR
Average Recall@IoU:0.5-0.95
30.8
best: 45.4 (Swin-T (ImageNet-1k pretrain))
2D ClassificationonTBBR
Average Recall@IoU:0.5-0.95
30.8
best: 45.4 (Swin-T (ImageNet-1k pretrain))
2D Object DetectiononTBBR
Average Recall@IoU:0.5-0.95
30.8
best: 45.4 (Swin-T (ImageNet-1k pretrain))
16konTBBR
Average Recall@IoU:0.5-0.95
30.8
best: 45.4 (Swin-T (ImageNet-1k pretrain))

Image ClassificationonCOCO-WAN
mIOU· 2024-06-16
25.5
SOTA
Benchmarking Label Noise in Instance Segmentation: Spatial Noise Matters arXiv:2406.10891
Object DetectiononCOCO minival
box AP· 2017-03-20
37.7
best: 66 (PE_spatial (DETA))
Mask R-CNN arXiv:1703.06870
Object DetectiononTBBR
Average Recall@IoU:0.5-0.95
30.8
best: 45.4 (Swin-T (ImageNet-1k pretrain))
Instance SegmentationonTBBR
Average Recall@IoU:0.5-0.95
20.1
best: 28 (Swin-T (ImageNet-1k pretrain))