Sparse Instance Activation for Real-Time Instance Segmentation

In this paper, we propose a conceptually novel, efficient, and fully convolutional framework for real-time instance segmentation. Previously, most instance segmentation methods heavily rely on object detection and perform mask prediction based on bounding boxes or dense centers. In contrast, we propose a sparse set of instance activation maps, as a new object representation, to highlight informative regions for each foreground object. Then instance-level features are obtained by aggregating features according to the highlighted regions for recognition and segmentation. Moreover, based on bipartite matching, the instance activation maps can predict objects in a one-to-one style, thus avoiding non-maximum suppression (NMS) in post-processing. Owing to the simple yet effective designs with instance activation maps, SparseInst has extremely fast inference speed and achieves 40 FPS and 37.9 AP on the COCO benchmark, which significantly outperforms the counterparts in terms of speed and accuracy. Code and models are available at https://github.com/hustvl/SparseInst.

PDF Abstract CVPR 2022 PDF CVPR 2022 Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Real-time Instance Segmentation MSCOCO SparseInst-608 (ResNet-50-vd) Frame (fps) 40 (2080 Ti) # 4
mask AP 37.9 # 1
AP50 59.2 # 2
AP75 40.2 # 1
APS 15.7 # 2
APM 39.4 # 2
APL 56.9 # 2
Real-time Instance Segmentation MSCOCO SparseInst-448 (ResNet-50-vd) Frame (fps) 58.5 (2080 Ti) # 1
mask AP 35.9 # 4
AP50 56.5 # 4
AP75 37.7 # 4
APS 12.3 # 5
APM 37.1 # 5
APL 57.0 # 1

Methods


No methods listed for this paper. Add relevant methods here