Figure 11From: Spatiotemporal Region Enhancement and Merging for Unsupervized Object SegmentationFrames 1, 90, 120, and 150 of Miss America sequence (no GM). Original clipProposed methodReference method 1 [1]Reference method 2 [2]Back to article page