Hierarchical Saliency Detection on Extended CSSD
This addresses a fundamental challenge in computer vision for applications like image segmentation and object recognition, but it appears incremental as it builds on existing scale-based methods.
The paper tackles the problem of saliency detection in images with small-scale high-contrast patterns, which causes errors in prior methods, by proposing a multi-layer approach that measures region-based scales and uses hierarchical inference to combine saliency cues, resulting in improved detection quality on many traditionally challenging images.
Complex structures commonly exist in natural images. When an image contains small-scale high-contrast patterns either in the background or foreground, saliency detection could be adversely affected, resulting erroneous and non-uniform saliency assignment. The issue forms a fundamental challenge for prior methods. We tackle it from a scale point of view and propose a multi-layer approach to analyze saliency cues. Different from varying patch sizes or downsizing images, we measure region-based scales. The final saliency values are inferred optimally combining all the saliency cues in different scales using hierarchical inference. Through our inference model, single-scale information is selected to obtain a saliency map. Our method improves detection quality on many images that cannot be handled well traditionally. We also construct an extended Complex Scene Saliency Dataset (ECSSD) to include complex but general natural images.