论文标题
学习预测语义细分的上下文自适应卷积
Learning to Predict Context-adaptive Convolution for Semantic Segmentation
论文作者
论文摘要
远程上下文信息对于实现高性能语义分割至关重要。先前的功能重新加权方法表明，使用全局上下文来重新加权特征通道可以有效提高语义分割的准确性。但是，全球共享功能重新加权向量可能对输入图像中不同类别的区域的区域可能不是最佳的。在本文中，我们提出了一个上下文自适应卷积网络（CAC-NET），以预测语义特征图的每个空间位置的空间变化特征加权向量。在CAC-NET中，以参数有效的方式从全局上下文信息中预测了一组上下文自适应卷积内核。当用于使用语义特征图的卷积时，预测的卷积内核可以生成捕获全球和局部上下文信息的空间变化特征权重因子。全面的实验结果表明，我们的CAC-NET在三个公共数据集（Pascal Context，Pascal VOC 2012和ADE20K）上实现了出色的细分性能。
Long-range contextual information is essential for achieving high-performance semantic segmentation. Previous feature re-weighting methods demonstrate that using global context for re-weighting feature channels can effectively improve the accuracy of semantic segmentation. However, the globally-sharing feature re-weighting vector might not be optimal for regions of different classes in the input image. In this paper, we propose a Context-adaptive Convolution Network (CaC-Net) to predict a spatially-varying feature weighting vector for each spatial location of the semantic feature maps. In CaC-Net, a set of context-adaptive convolution kernels are predicted from the global contextual information in a parameter-efficient manner. When used for convolution with the semantic feature maps, the predicted convolutional kernels can generate the spatially-varying feature weighting factors capturing both global and local contextual information. Comprehensive experimental results show that our CaC-Net achieves superior segmentation performance on three public datasets, PASCAL Context, PASCAL VOC 2012 and ADE20K.
