M-CenterNet (DLA-34)

Reported on 35 benchmarks across 5 tasks

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

Methodology28 results

3DonAI-TOD
AP
14.5
best: 26.2 (DNTR)
3DonAI-TOD
AP50
40.7
best: 56.7 (DNTR)
3DonAI-TOD
AP75
6.4
best: 20.2 (DNTR)
3DonAI-TOD
APm
20.4
best: 38.3 (DetectoRS + NWD (ResNet-50-FPN))
3DonAI-TOD
APs
19.4
best: 31 (DNTR)
3DonAI-TOD
APt
15
best: 26.4 (DNTR)
3DonAI-TOD
APvt
6.1
best: 12.8 (DNTR)
2D ClassificationonAI-TOD
AP
14.5
best: 26.2 (DNTR)
2D ClassificationonAI-TOD
AP50
40.7
best: 56.7 (DNTR)
2D ClassificationonAI-TOD
AP75
6.4
best: 20.2 (DNTR)
2D ClassificationonAI-TOD
APm
20.4
best: 38.3 (DetectoRS + NWD (ResNet-50-FPN))
2D ClassificationonAI-TOD
APs
19.4
best: 31 (DNTR)
2D ClassificationonAI-TOD
APt
15
best: 26.4 (DNTR)
2D ClassificationonAI-TOD
APvt
6.1
best: 12.8 (DNTR)
2D Object DetectiononAI-TOD
AP
14.5
best: 26.2 (DNTR)
2D Object DetectiononAI-TOD
AP50
40.7
best: 56.7 (DNTR)
2D Object DetectiononAI-TOD
AP75
6.4
best: 20.2 (DNTR)
2D Object DetectiononAI-TOD
APm
20.4
best: 38.3 (DetectoRS + NWD (ResNet-50-FPN))
2D Object DetectiononAI-TOD
APs
19.4
best: 31 (DNTR)
2D Object DetectiononAI-TOD
APt
15
best: 26.4 (DNTR)
2D Object DetectiononAI-TOD
APvt
6.1
best: 12.8 (DNTR)
16konAI-TOD
AP
14.5
best: 26.2 (DNTR)
16konAI-TOD
AP50
40.7
best: 56.7 (DNTR)
16konAI-TOD
AP75
6.4
best: 20.2 (DNTR)
16konAI-TOD
APm
20.4
best: 38.3 (DetectoRS + NWD (ResNet-50-FPN))
16konAI-TOD
APs
19.4
best: 31 (DNTR)
16konAI-TOD
APt
15
best: 26.4 (DNTR)
16konAI-TOD
APvt
6.1
best: 12.8 (DNTR)