Superpixel Level Incremental Visual Saliency Detection in Low Contrast Video
Zhen Zhang, Xin Xu, Xiaowei Fu · 2017
The last few decades have witnessed the rapid development of saliency detection, which can automatically extract object-of-interest from clutter scene. However, visual saliency detection in low contrast video stream still remains a challenge. In this paper, we present a saliency detection model to detect salient object in low contrast video, which combines spatial saliency computation with temporal saliency computation. In spatial domain, superpixel segmentation and boundary prior are utilized to detect salient object in single video frame. And incremental learning algorithm is employed in temporal domain to effectively update the background model. Extensive experiments demonstrate that this method can achieve better performance.