Deep Variational Instance Segmentation
Deep Variational Instance Segmentation
复制标题
DOI：
--
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Jialing Yuan;Chao Chen;Fuxin Li
中科院分区：
文献类型：
--
作者：
Jialing Yuan;Chao Chen;Fuxin Li
实例分割是计算机视觉中的一项具有挑战性的任务，它试图为输入图像中的每个像素获得类别和实例标签。最先进的算法通常采用两个独立的阶段，第一个阶段生成对象建议，第二个阶段识别和细化边界。此外，建议通常基于更快的R-CNN等检测器，它们在整个图像中彻底搜索框。在本文中，我们提出了一种新的算法，直接利用全卷积网络（FCN）来预测实例标签。具体来说，我们提出了一个变分松弛的实例分割最小化的分段常数分割问题，它可以用来训练一个FCN端到端的优化功能。它扩展了经典的Mumford-Shah变分分割问题，使其能够处理实例分割的地面真值中的置换不变标签。在PASCAL VOC 2012、语义边界数据集（SBD）和MSCOCO 2017数据集上的实验表明，该方法有效地解决了实例分割任务。源代码和训练模型将与论文一起发布。
Instance Segmentation, which seeks to obtain both class and instance labels for each pixel in the input image, is a challenging task in computer vision. State-of-the-art algorithms often employ two separate stages, the first one generating object proposals and the second one recognizing and refining the boundaries. Further, proposals are usually based on detectors such as faster R-CNN which search for boxes in the entire image exhaustively. In this paper, we propose a novel algorithm that directly utilizes a fully convolutional network (FCN) to predict instance labels. Specifically, we propose a variational relaxation of instance segmentation as minimizing an optimization functional for a piecewise-constant segmentation problem, which can be used to train an FCN end-to-end. It extends the classical Mumford-Shah variational segmentation problem to be able to handle permutation-invariant labels in the ground truth of instance segmentation. Experiments on PASCAL VOC 2012, Semantic Boundaries dataset(SBD), and the MSCOCO 2017 dataset show that the proposed approach efficiently tackle the instance segmentation task. The source code and trained models will be released with the paper.