单幅图片法向量估计是计算机图形学和计算机视觉研究的重要问题之一． 在缺 少其它三维信息的情况下，由单幅图像预测出对应法向量，对于三维场景重建，三维模型 识别，三维语义分割等具有重要意义． 为解决这一问题，文中使用多尺度的卷积网络结构， 对图像进行端到端的输出预测． 该网络由两个层级组成，第 1 层采用在 ImageNet 中性能 最好的 DenseNet 分类网络，对输入进行全局处理． 第 2 层级采用全卷积网络结构，对第 1 层级获得的输出进行进一步的精细预测． 实验结果表明，即使不使用其他预处理或后处理 步骤，文中提出的网络在单幅图像点法向预测方面仍能取得较理想的结果．

Normal estimation from monocular images is one of the most important issues in computer graphics and computer vision research． Short of three-dimensional information，the corresponding normal is predicted from the monocular images，which is of great significance for 3D scene reconstruction，3D model recognition，3D semantic segmentation，etc． In order to find the solution to the problem，this paper adopts a multi-scale convolutional net- work structure to predict an end-to-end output of the image． The network consists of two scales，the first layer uses the DenseNet classification network with the best performance in ImageNet to process the input globally． The second level uses a fully convolutional network to further fine-tune the output obtained from the first level． The experimen- tal results show that the network proposed in this paper can achieve better results in normal prediction of monocular image even without using other pre-processing or post-processing steps．