VarifocalNet is a method aimed at accurately ranking a huge number of candidate detections in object detection. It consists of a new loss function, named Varifocal Loss, for training a dense object detector to predict the IACS, and a new efficient star-shaped bounding box feature representation for estimating the IACS and refining coarse bounding boxes. Combining these two new components and a bounding box refinement branch, results in a dense object detector on the FCOS architecture, what the authors call VarifocalNet or VFNet for short.
Source: VarifocalNet: An IoU-aware Dense Object DetectorPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Object Detection | 4 | 50.00% |
Small Object Detection | 1 | 12.50% |
Instance Segmentation | 1 | 12.50% |
Semantic Segmentation | 1 | 12.50% |
General Classification | 1 | 12.50% |