FCOS is an anchor-box free, proposal free, single-stage object detection model. By eliminating the predefined set of anchor boxes, FCOS avoids computation related to anchor boxes such as calculating overlapping during training. It also avoids all hyper-parameters related to anchor boxes, which are often very sensitive to the final detection performance.
Source: FCOS: Fully Convolutional One-Stage Object DetectionPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Object Detection | 55 | 35.71% |
Semantic Segmentation | 13 | 8.44% |
Instance Segmentation | 12 | 7.79% |
Pedestrian Detection | 8 | 5.19% |
Pseudo Label | 4 | 2.60% |
Autonomous Driving | 4 | 2.60% |
Domain Adaptation | 3 | 1.95% |
Pose Estimation | 3 | 1.95% |
Decoder | 3 | 1.95% |