HyperAI

Semi Supervised Object Detection On Coco 5

Metriken

mAP

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Vergleichstabelle
ModellnamemAP
semi-supervised-object-detection-with-131.35±0.13
adaptive-self-training-for-object-detection30.43
pseco-pseudo-labeling-and-consistency32.5
omni-detr-omni-supervised-object-detection30.2
semi-detr-semi-supervised-object-detection-140.1
detreg-unsupervised-pretraining-with-region24.80±0.2
mixed-pseudo-labels-for-semi-supervised40.1
mind-the-gap-polishing-pseudo-labels-for32.10±0.15
revisiting-class-imbalance-for-end-to-end32.21
dense-teacher-dense-pseudo-labels-for-semi33.01
mixteacher-mining-promising-labels-with-mixed33.42
ambiguity-resistant-semi-supervised-learning34.45
mum-mix-image-tiles-and-unmix-feature-tiles28.52
unbiased-teacher-v2-semi-supervised-object-131.85±0.09
unbiased-teacher-for-semi-supervised-object-128.27± 0.11
mixteacher-mining-promising-labels-with-mixed34.06
efficient-teacher-semi-supervised-object34.11
rethinking-pseudo-labels-for-semi-supervised28.4 ± 0.15
a-simple-semi-supervised-learning-framework24.38±0.12
semi-supervised-object-detection-via-virtual-132.05
instant-teaching-an-end-to-end-semi26.75
semi-supervised-object-detection-with-object30.66
end-to-end-semi-supervised-object-detection30.74