Recently, although deep learning network has shown its advantages in supervised salient object detection, supervised models often require massive pixel-wise annotations and learnable parameters, which seriously manacle training and testing of models. In this paper, we present a mix-supervised unified framework for salient object detection to avoid the insufficient training labels and speed training and testing up, which is composed of a region-wise stream and a pixel-wise stream. In the region-wise stream, to avoid the requirement of expensive pixel-wise annotations, an improved energy equation based manifold learning algorithm is employed, by which accurate object location and prior knowledge are introduced by the unsupervised learning. In the pixel-wise stream, to alleviate the problem of time-consuming, a simplified bi-directional reuse network is introduced, which can obtain clear object contour and competitive performance with fewer parameters. To relieve the bottleneck pressure of parallel training and testing, each steam is directly connected to its pre-processed color feature and post-processing refinement. Extensive experiments demonstrate that each component contributes to the final results and complement each other perfectly.

最近，尽管深度学习网络在有监督的显着对象检测中已显示出其优势，但受监督的模型通常需要大量的逐像素注释和可学习的参数，这严重地破坏了模型的训练和测试。在本文中，我们提出了一个混合监督的统一框架，用于显着目标检测，以避免训练标签不足和速度训练和测试，它由区域流和像素流组成。在区域流中，为避免昂贵的逐像素注释，采用了一种基于改进的能量方程的流形学习算法，通过无监督学习引入了精确的对象位置和先​​验知识。在像素级流中，为了减轻耗时的问题，引入了简化的双向重用网络，该网络可以使用较少的参数获得清晰的对象轮廓和竞争性能。为了缓解并行训练和测试的瓶颈压力，每个蒸汽都直接连接到其预处理的颜色特征和后期处理的精炼上。大量的实验表明，每种成分都有助于最终结果并彼此完美互补。