What are anchor boxes and their downsides?
mediumAnswer
- Predefined reference boxes of various sizes and aspect ratios placed at every location of the feature map.
- The network predicts offsets and class scores relative to each anchor.
- Downsides: (1) requires careful anchor design (scales, ratios) per dataset; (2) massive class imbalance (foreground:background can be 1:1000); (3) many hyperparameters.
- Anchor-free methods (FCOS, CenterNet, DETR) predict box centers or points directly and skip anchors.
Check yourself — multiple choice
- Anchors are always necessary
- Predefined boxes at every location; downsides: design overhead, imbalance, hyperparams — FCOS/CenterNet/DETR skip them
- Anchors are learned
- No downsides
Anchors: predefined reference boxes; anchor-free methods sidestep design + imbalance.
#detection#computer-vision
Practise Deep Learning
214 interview questions in this topic.
Related questions
- How does a two-stage detector (Faster R-CNN) work?
- How does YOLO / SSD / RetinaNet differ from two-stage detection?
- What is Non-Maximum Suppression (NMS) and its variants?
- How does Mask R-CNN extend Faster R-CNN for instance segmentation?
- What is a Feature Pyramid Network (FPN)?
- How does DETR reformulate object detection?