Pixelwise Instance Segmentation with a Dynamically Instantiated Network
Pixelwise Instance Segmentation with a Dynamically Instantiated Network
复制标题
DOI：
10.1109/cvpr.2017.100
复制
发表时间：
2017-04
期刊：
影响因子：
--
通讯作者：
Anurag Arnab;Philip H. S. Torr
中科院分区：
文献类型：
--
作者：
Anurag Arnab;Philip H. S. Torr
语义分割和目标检测研究近期取得了快速进展。然而，前者的任务没有同一物体不同实例的概念，而后者是在粗糙的边界框级别上进行操作。我们提出了一种实例分割系统，它能生成一个分割图，其中每个像素都被赋予一个物体类别和实例身份标签。大多数方法是调整目标探测器来生成分割区域而不是边界框。相比之下，我们的方法基于一个初始的语义分割模块，该模块为一个实例子网络提供输入。这个子网络在一个端到端的条件随机场（CRF）中，利用初始的类别级分割以及来自目标探测器输出的线索来预测实例。我们模型的这一部分是动态实例化的，以便为每张图像生成数量可变的实例。我们的端到端方法不需要后处理，并且从整体上考虑图像，而不是处理独立的候选区域。因此，与一些相关工作不同，一个像素不能属于多个实例。此外，正如我们在高平均精度（APr）阈值下的显著改进所表明的那样，我们实现了更精确的分割。
Semantic segmentation and object detection research have recently achieved rapid progress. However, the former task has no notion of different instances of the same object, and the latter operates at a coarse, bounding-box level. We propose an Instance Segmentation system that produces a segmentation map where each pixel is assigned an object class and instance identity label. Most approaches adapt object detectors to produce segments instead of boxes. In contrast, our method is based on an initial semantic segmentation module, which feeds into an instance subnetwork. This subnetwork uses the initial category-level segmentation, along with cues from the output of an object detector, within an end-to-end CRF to predict instances. This part of our model is dynamically instantiated to produce a variable number of instances per image. Our end-to-end approach requires no post-processing and considers the image holistically, instead of processing independent proposals. Therefore, unlike some related work, a pixel cannot belong to multiple instances. Furthermore, far more precise segmentations are achieved, as shown by our substantial improvements at high APr thresholds.