Mask R-CNN

A two-stage detection network extended with a branch that predicts a segmentation mask for each detected object.

Mask R-CNN extends a two-stage object detector (propose candidate regions, then classify and refine each one) with an added mask branch that predicts a per-pixel segmentation mask inside each detected box, producing instance segmentation as detection plus a mask per detected instance.

Examples

Mask R-CNN first proposes candidate person-shaped regions, then predicts a per-pixel mask inside each proposed box.