Question about the paper

[wuzhenyubuaa]

The author claims in their paper that their method is unsupervised method. Actually, as in stated in the paper, the author use clean GT as supervision. if I miss something, could you explain it?

[Kinpzz]

We define our approach as an unsupervised domain adaptive salient object detection method, but not an unsupervised method. Only when comparing with others, we follow the definition of existing deep USOD methods[1,2] that defines "unsupervised learning" as learning without human annotations. Most of them refer to models trained with noise labels generated by traditional methods.

References [1] Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective, CVPR 2018 [2] DeepUSPS: Deep Robust Unsupervised Saliency Prediction With Self-Supervision, NeurIPS 2019

[wuzhenyubuaa]

Thanks for your reply! I stand with your response. There are some misleading sentence in the conclusion----"In this paper, we propose to tackle deep unsupervised salient object detection from a novel perspective, i.e., learning from synthetic but clean labels"

[Kinpzz]

Thanks for pointing it out. Actually, we want to express that our approach learns from synthetic but clean labels instead of noise labels generated by traditional methods like existing methods. Sorry for the misleading and we will add this question to FAQ section of the release code for better explanation.

[wangjinkai8611]

您好，请问下合成图像的标签是怎么得到的？

[Kinpzz]

@wangjinkai8611 网上搜集的带有透明背景的素材图片的Alpha通道