We propose a hybrid architecture composed of a fully convolutional network (FCN) and a Dempster-Shafer layer for image semantic segmentation. In the so-called evidential FCN (E-FCN), an encoder-decoder architecture first extracts pixel-wise feature maps from an input image. A Dempster-Shafer layer then computes mass functions at each pixel location based on distances to prototypes. Finally, a utility layer performs semantic segmentation from mass functions and allows for imprecise classification of ambiguous pixels and outliers. We propose an end-to-end learning strategy for jointly updating the network parameters, which can make use of soft (imprecise) labels. Experiments using three databases (Pascal VOC 2011, MIT-scene Parsing and SIFT Flow) show that the proposed combination improves the accuracy and calibration of semantic segmentation by assigning confusing pixels to multi-class sets.

翻译：我们建议建立一个混合结构,由完全进化网络和Dempster-Shafer层组成,用于图像语义分解。在所谓的证据 FCN(E-FCN)中,一个编码器解码器结构首先从输入图像中提取像素特征图。一个Dempster-Shafer层然后根据离原型的距离计算每个像素位置的质量函数。最后,一个公用层从质量函数中进行语义分解,并允许对模糊的像素和离子进行不精确的分类。我们提出了联合更新网络参数的端对端学习战略,这可以使用软(简化)标签。使用三个数据库的实验(Pascal VOC,2011年,MIT-cene 剖析和SIFT流)表明,拟议的组合通过将混淆像素分解到多级组,提高了语义分解的准确度和校准。