Recent advances in salient object detection -- Towards object recognition in big media data
Research output: Contribution to journal › Article › Scientific › peer-review
|Number of pages||92|
|Publication status||Published - 2016|
|Publication type||A1 Journal article-refereed|
With the massive size of today’s image collections, challenges related to the ability of image analysis approaches to operate in Big Media Data have been raised. Salient object detection is an emerging research topic, which has been proposed as a way of reducing the complexity of subsequent analysis tasks, such as object and scene recognition, by highlighting salient image locations that can be used for further analysis in subsequent steps. In this paper, we provide an extensive survey of salient object detection approaches. We start with a theoretical analysis of eight state-of-the-art salient object detection methods. These methods are then qualitatively compared in terms of the saliency cues they exploit and their performance on benchmark images. Publicly available salient object detection benchmark datasets and the commonly used evaluation metrics are also described and discussed.
- attention mechanism, unsupervised image analysis, Salient Object Segmentation, visual saliency