1 paper
Shuyong Gao, Wei Zhang, Yan Wang +4
Current state-of-the-art saliency detection models rely heavily on large datasets of accurate pixel-wise annotations, but manually labeling pixels is time-consuming and labor-inten…