Mask R-CNN

ICCV 2017 Kaiming HeGeorgia GkioxariPiotr DollárRoss Girshick

We present a conceptually simple, flexible, and general framework for object instance segmentation. Our approach efficiently detects objects in an image while simultaneously generating a high-quality segmentation mask for each instance... (read more)

PDF Abstract

Code


Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK USES EXTRA
TRAINING DATA
COMPARE
Nuclear Segmentation Cell17 Mask R-CNN F1-score 0.8004 # 3
Nuclear Segmentation Cell17 Mask R-CNN Dice 0.7070 # 3
Nuclear Segmentation Cell17 Mask R-CNN Hausdorff 12.6723 # 3
Instance Segmentation Cityscapes test Mask R-CNN + Coco Average Precision 32.0 # 3
Instance Segmentation Cityscapes test Mask R-CNN Average Precision 26.2 # 4
Keypoint Detection COCO Mask R-CNN Validation AP 69.2 # 6
Keypoint Detection COCO Mask R-CNN Test AP 63.1 # 9
Instance Segmentation COCO minival Mask R-CNN (ResNet-101-FPN) mask AP 35.4 # 14
Object Detection COCO minival Mask R-CNN (ResNeXt-101-FPN) box AP 36.7 # 48
Object Detection COCO minival Mask R-CNN (ResNeXt-101-FPN) AP50 59.5 # 20
Object Detection COCO minival Mask R-CNN (ResNeXt-101-FPN) AP75 38.9 # 31
Object Detection COCO minival Mask R-CNN (ResNet-101-FPN) box AP 40.0 # 36
Object Detection COCO minival Mask R-CNN (ResNet-50-FPN) box AP 37.7 # 47
Instance Segmentation COCO minival Mask R-CNN (ResNet-50-FPN) mask AP 33.6 # 15
Instance Segmentation COCO minival Mask R-CNN (ResNeXt-101-FPN) mask AP 36.7 # 11
Instance Segmentation COCO test-dev Mask R-CNN (ResNeXt-101-FPN) mask AP 37.1% # 14
Instance Segmentation COCO test-dev Mask R-CNN (ResNeXt-101-FPN) AP50 60 # 7
Instance Segmentation COCO test-dev Mask R-CNN (ResNeXt-101-FPN) AP75 39.4 # 7
Instance Segmentation COCO test-dev Mask R-CNN (ResNeXt-101-FPN) APS 16.9 # 8
Instance Segmentation COCO test-dev Mask R-CNN (ResNeXt-101-FPN) APM 39.9 # 8
Instance Segmentation COCO test-dev Mask R-CNN (ResNeXt-101-FPN) APL 53.5 # 7
Object Detection COCO test-dev Mask R-CNN (ResNeXt-101-FPN) box AP 39.8 # 52
Object Detection COCO test-dev Mask R-CNN (ResNeXt-101-FPN) AP50 62.3 # 33
Object Detection COCO test-dev Mask R-CNN (ResNeXt-101-FPN) AP75 43.4 # 44
Object Detection COCO test-dev Mask R-CNN (ResNeXt-101-FPN) APS 22.1 # 46
Object Detection COCO test-dev Mask R-CNN (ResNeXt-101-FPN) APM 43.2 # 39
Object Detection COCO test-dev Mask R-CNN (ResNeXt-101-FPN) APL 51.2 # 46
Object Detection COCO test-dev Mask R-CNN (ResNet-101-FPN) box AP 38.2 # 60
Object Detection COCO test-dev Mask R-CNN (ResNet-101-FPN) AP50 60.3 # 42
Object Detection COCO test-dev Mask R-CNN (ResNet-101-FPN) AP75 41.7 # 53
Object Detection COCO test-dev Mask R-CNN (ResNet-101-FPN) APS 20.1 # 57
Object Detection COCO test-dev Mask R-CNN (ResNet-101-FPN) APM 41.1 # 48
Object Detection COCO test-dev Mask R-CNN (ResNet-101-FPN) APL 50.2 # 49
Instance Segmentation COCO test-dev Mask R-CNN (ResNet-101-FPN) mask AP 35.7% # 16
Multi-Human Parsing MHP v1.0 Mask R-CNN AP 0.5 52.68% # 2
Multi-Human Parsing MHP v2.0 Mask R-CNN AP 0.5 14.90% # 3