EasyDeepLearn

How does Mask R-CNN extend Faster R-CNN for instance segmentation?

hard

Answer

  • Add a parallel FCN mask-prediction head to each RoI, predicting a small binary mask (28×28 typically) per class.
  • Replaces RoIPool with RoIAlign (bilinear interpolation instead of quantization) — critical for precise mask localization.
  • Preserves the classification / box head.
  • Standard instance-segmentation baseline; adopted broadly for pose estimation, keypoints, panoptic segmentation.
Check yourself — multiple choice
  • Same as U-Net
  • Faster R-CNN + per-RoI mask FCN head + RoIAlign → instance segmentation
  • No mask head
  • Removes classification

Mask R-CNN: Faster R-CNN + mask FCN + RoIAlign → instance seg.

#segmentation#computer-vision#detection

Practise Deep Learning

214 interview questions in this topic.

Related questions