CenterMask is an anchor-free instance segmentation method that adds a novel spatial attention-guided mask (SAG-Mask) branch to anchor-free one stage object detector (FCOS) in the same vein with Mask R-CNN. Plugged into the FCOS object detector, the SAG-Mask branch predicts a segmentation mask on each detected box with the spatial attention map that helps to focus on informative pixels and suppress noise.
Source: CenterMask : Real-Time Anchor-Free Instance SegmentationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Instance Segmentation | 3 | 21.43% |
Semantic Segmentation | 3 | 21.43% |
Semi-Supervised Instance Segmentation | 2 | 14.29% |
Pseudo Label | 1 | 7.14% |
Defect Detection | 1 | 7.14% |
Object Detection | 1 | 7.14% |
Panoptic Segmentation | 1 | 7.14% |
Real-time Instance Segmentation | 1 | 7.14% |
Real-Time Object Detection | 1 | 7.14% |
Component | Type |
|
---|---|---|
FCOS
|
Object Detection Models | |
FPN
|
Feature Extractors | |
ReLU
|
Activation Functions | |
RoIAlign
|
RoI Feature Extractors | |
Spatial Attention-Guided Mask
|
Mask Branches |