Active Boundary Loss for Semantic Segmentation

This paper proposes a novel active boundary loss for semantic segmentation. It can progressively encourage the alignment between predicted boundaries and ground-truth during end-to-end training, which is not explicitly enforced in commonly used cross-entropy loss. Based on detected from segmentation results using current network parameters, we formulate problem as differentiable direction vector prediction to guide movement of each iteration. Our model-agnostic be plugged training networks improve details. Experimental show that with effectively F-score mean Intersection-over-Union challenging image video object datasets.