Liver tumor segmentation plays an important role in hepatocellular carcinoma diagnosis and surgical planning. In this paper, we propose a novel Semantic Feature Attention Network (SFAN) for liver tumor segmentation from Computed Tomography (CT) volumes, which exploits the impact of both low-level and high-level features. In the SFAN, a Semantic Attention Transmission (SAT) module is designed to select discriminative low-level localization details with the guidance of neighboring high-level semantic information. Furthermore, a Global Context Attention (GCA) module is proposed to effectively fuse the multi-level features with the guidance of global context. Our experiments are based on 2 challenging databases, the public Liver Tumor Segmentation (LiTS) Challenge database and a large-scale in-house clinical database with 912 CT volumes. Experimental results show that our proposed framework can not only achieve the state-of-the-art performance with the Dice per case on liver tumor segmentation in LiTS database, but also outperform some widely used segmentation algorithms in the large-scale clinical database.