Unsupervised Domain Adaptive Salient Object Detection Through Uncertainty-Aware Pseudo-Label Learning
This work addresses the labeling burden in salient object detection for computer vision applications, representing an incremental improvement by adapting synthetic data to real-world domains.
The paper tackles the problem of reducing the need for labor-intensive pixel-level annotations in salient object detection by proposing an unsupervised domain adaptation method that uses synthetic data with clean labels and adapts to real-world scenarios through uncertainty-aware self-training. The result is that the method outperforms existing state-of-the-art unsupervised methods on benchmark datasets and achieves performance comparable to fully-supervised approaches.
Recent advances in deep learning significantly boost the performance of salient object detection (SOD) at the expense of labeling larger-scale per-pixel annotations. To relieve the burden of labor-intensive labeling, deep unsupervised SOD methods have been proposed to exploit noisy labels generated by handcrafted saliency methods. However, it is still difficult to learn accurate saliency details from rough noisy labels. In this paper, we propose to learn saliency from synthetic but clean labels, which naturally has higher pixel-labeling quality without the effort of manual annotations. Specifically, we first construct a novel synthetic SOD dataset by a simple copy-paste strategy. Considering the large appearance differences between the synthetic and real-world scenarios, directly training with synthetic data will lead to performance degradation on real-world scenarios. To mitigate this problem, we propose a novel unsupervised domain adaptive SOD method to adapt between these two domains by uncertainty-aware self-training. Experimental results show that our proposed method outperforms the existing state-of-the-art deep unsupervised SOD methods on several benchmark datasets, and is even comparable to fully-supervised ones.