Path Aggregation Network for Instance Segmentation

CVPR 2018  ·  Shu Liu, Lu Qi, Haifang Qin, Jianping Shi, Jiaya Jia ·

The way that information propagates in neural networks is of great importance. In this paper, we propose Path Aggregation Network (PANet) aiming at boosting information flow in proposal-based instance segmentation framework. Specifically, we enhance the entire feature hierarchy with accurate localization signals in lower layers by bottom-up path augmentation, which shortens the information path between lower layers and topmost feature. We present adaptive feature pooling, which links feature grid and all feature levels to make useful information in each feature level propagate directly to following proposal subnetworks. A complementary branch capturing different views for each proposal is created to further improve mask prediction. These improvements are simple to implement, with subtle extra computational overhead. Our PANet reaches the 1st place in the COCO 2017 Challenge Instance Segmentation task and the 2nd place in Object Detection task without large-batch training. It is also state-of-the-art on MVD and Cityscapes. Code is available at https://github.com/ShuLiu1993/PANet

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Instance Segmentation COCO minival PANet (ResNet-50) mask AP 37.8 # 80
Object Detection COCO test-dev PANet (ResNeXt-101, multi-scale) box mAP 47.4 # 112
AP50 67.2 # 63
AP75 51.8 # 60
APS 30.1 # 49
APM 51.7 # 49
APL 60.0 # 57
Hardware Burden None # 1
Operations per network pass None # 1
Instance Segmentation COCO test-dev PANet mask AP 42.0 # 54
Object Detection iSAID PANet Average Precision 41.66 # 3
Instance Segmentation iSAID PANet Average Precision 34.17 # 5

Methods