BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation

CVPR 2020 Hao ChenKunyang SunZhi TianChunhua ShenYongming HuangYouliang Yan

Instance segmentation is one of the fundamental vision tasks. Recently, fully convolutional instance segmentation methods have drawn much attention as they are often simpler and more efficient than two-stage approaches like Mask R-CNN... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Instance Segmentation COCO test-dev BlendMask (ResNet-101 + DCN interval=3) mask AP 41.3 # 8
AP50 63.1 # 6
AP75 44.6 # 6
APS 22.7 # 6
APM 44.1 # 6
APL 54.5 # 7
Real-time Instance Segmentation MSCOCO BlendMask-512 (DLA_34) AP 35.2 # 5
Frame (fps) 33.3 # 3

Methods used in the Paper