Weakly Supervised Video Salient Object Detection
This work addresses the time-consuming and expensive data annotation issue in video salient object detection, offering a practical solution for researchers and practitioners in computer vision.
The paper tackles the problem of reducing annotation costs for video salient object detection by introducing the first weakly supervised model using relabeled fixation guided scribble annotations, achieving effective performance on six benchmark datasets.
Significant performance improvement has been achieved for fully-supervised video salient object detection with the pixel-wise labeled training datasets, which are time-consuming and expensive to obtain. To relieve the burden of data annotation, we present the first weakly supervised video salient object detection model based on relabeled "fixation guided scribble annotations". Specifically, an "Appearance-motion fusion module" and bidirectional ConvLSTM based framework are proposed to achieve effective multi-modal learning and long-term temporal context modeling based on our new weak annotations. Further, we design a novel foreground-background similarity loss to further explore the labeling similarity across frames. A weak annotation boosting strategy is also introduced to boost our model performance with a new pseudo-label generation technique. Extensive experimental results on six benchmark video saliency detection datasets illustrate the effectiveness of our solution.