The Secrets of Salient Object Segmentation
It addresses flaws in salient object segmentation benchmarks for computer vision researchers, though it is incremental in improving dataset quality.
The paper identifies dataset design bias in existing salient object segmentation benchmarks and proposes a new high-quality dataset with both fixation and segmentation ground-truth, leading to significant benchmark progress on three datasets.
In this paper we provide an extensive evaluation of fixation prediction and salient object segmentation algorithms as well as statistics of major datasets. Our analysis identifies serious design flaws of existing salient object benchmarks, called the dataset design bias, by over emphasizing the stereotypical concepts of saliency. The dataset design bias does not only create the discomforting disconnection between fixations and salient object segmentation, but also misleads the algorithm designing. Based on our analysis, we propose a new high quality dataset that offers both fixation and salient object segmentation ground-truth. With fixations and salient object being presented simultaneously, we are able to bridge the gap between fixations and salient objects, and propose a novel method for salient object segmentation. Finally, we report significant benchmark progress on three existing datasets of segmenting salient objects