Salient Objects in Clutter: Bringing Salient Object Detection to the Foreground
This work addresses the problem of biased evaluation in salient object detection for computer vision researchers, but it is incremental as it focuses on dataset creation rather than a new method.
The paper identifies a design bias in existing salient object detection datasets, where models perform well on low-clutter images but poorly in real-world scenes, and introduces a new dataset (SOC) with attributes to address this, reporting attribute-based performance.
We provide a comprehensive evaluation of salient object detection (SOD) models. Our analysis identifies a serious design bias of existing SOD datasets which assumes that each image contains at least one clearly outstanding salient object in low clutter. The design bias has led to a saturated high performance for state-of-the-art SOD models when evaluated on existing datasets. The models, however, still perform far from being satisfactory when applied to real-world daily scenes. Based on our analyses, we first identify 7 crucial aspects that a comprehensive and balanced dataset should fulfill. Then, we propose a new high quality dataset and update the previous saliency benchmark. Specifically, our SOC (Salient Objects in Clutter) dataset, includes images with salient and non-salient objects from daily object categories. Beyond object category annotations, each salient image is accompanied by attributes that reflect common challenges in real-world scenes. Finally, we report attribute-based performance assessment on our dataset.