Fig. 5
From: Joint level-set and spatio-temporal motion detection for cell segmentation
Intermediate results produced by ST-Diff-TCV on sample frames of a C2DL-MSC02 and b N2DH-SIM04 data sequences. First row: center, previous and next frames in the temporal space (left to right) Second row: S-T diffused frame, kernel density estimation of edge-moving regions then the inverted probability density map. Third row: watershed result, cell identification after foreground/background separation, and the reference segmentation mask (left to right)