- symmetric:对称的
- We propose a symmetric encoder-decoder CNN architecture to extract global and multi-scale feature maps.
- as indicated by:如…所示
- The encoder-decoder network of Figure 1 uses a deep symmetric CNN architecture with skip connections as indicated by black arrows.
- objective & subjective:客观的 & 主观的
- This part is similar to U-Net, but the difference is that U-net is designed for image segmentation, which is objective and works well even with cropped feature maps. For saliency segmentation, it is subjective and easily affected in different scenarios.
- ultimately:最终
- We ultimately want to distinguish salient objects from background and so want to map image pixels into a feature space where that distance across salient and background regions is large, but within regions is small.
- endeavor:努力
- As the CNN becomes increasingly deeper, recent works endeavor to refine or reuse the features from previous layers through identity mappings or concatenation.
- diverge:分歧
- The CNNs designed for image-level, region-level, and pixel-level tasks begin to diverge in network structure.
- diversity:多样性
- All of them should be kept to improve the diversity of features.
- symbolize:象征着,标志着
- The remarkable improvement in the image recognition challenge ILSVRC achieved by AlexNet symbolizes a new era of deep learning for computer vision.
- is relieved by:被缓解
- Recently, the problem of vanishing gradient is greatly relieved by introducing the skip connections into the network.
- applicable:使用的
- Since the numbers of channels for and are different, identity mapping is not applicable.