Deep learning approaches to building rooftop thermal bridge detection from aerial images

Open paper
Benchmark Model Rank Results
instance-segmentation-on-tbbrSwin-T (ImageNet-1k pretrain)#1Average Recall@IoU:0.5-0.95: 28.0
instance-segmentation-on-tbbrMask R-CNN (ResNet-50-FPN, ImageNet-1k pretrain)#2Average Recall@IoU:0.5-0.95: 21.9
instance-segmentation-on-tbbrSwin-T#3Average Recall@IoU:0.5-0.95: 20.6
instance-segmentation-on-tbbrMask R-CNN (ResNet-50-FPN)#4Average Recall@IoU:0.5-0.95: 20.1
object-detection-on-tbbrSwin-T (ImageNet-1k pretrain)#1Average Recall@IoU:0.5-0.95: 45.4
object-detection-on-tbbrFSAF (ResNeXt-101, ImageNet-1k pretrain)#2Average Recall@IoU:0.5-0.95: 38.0
object-detection-on-tbbrMask R-CNN (ResNet-50-FPN, ImageNet-1k pretrain)#3Average Recall@IoU:0.5-0.95: 37.0
object-detection-on-tbbrMask R-CNN (ResNet-50-FPN)#4Average Recall@IoU:0.5-0.95: 30.8
object-detection-on-tbbrTridentNet (ResNet-50, ImageNet-1k pretrain)#5Average Recall@IoU:0.5-0.95: 30.0
object-detection-on-tbbrFSAF (ResNeXt-101)#6Average Recall@IoU:0.5-0.95: 24.8
object-detection-on-tbbrTridentNet (ResNet-50)#7Average Recall@IoU:0.5-0.95: 21.5

Rank counts only results with a code link.