Exemplar-Based Recursive Instance Segmentation With Application to Plant Image Analysis
Exemplar-Based Recursive Instance Segmentation With Application to Plant Image Analysis
复制标题
基于样本的递归实例分割在植物图像分析中的应用
DOI：
10.1109/tip.2019.2923571
复制
发表时间：
2020
影响因子：
10.6
通讯作者：
Yuanqing Li
中科院分区：
文献类型：
--
作者：
Jin-Gang Yu;Yansheng Li;Changxin Gao;Hongxia Gao;Gui-Song Xia;Zhu Liang Yu;Yuanqing Li
实例分割是一个具有挑战性的计算机视觉问题，它处于目标检测和语义分割的交叉点。植物表型分析是计算机视觉的一个新兴应用领域，本文针对植物图像分析提出了基于样本的递归实例分割（ERIS）框架。首先引入一个三层概率模型来联合表示假设、投票元素、实例标签及其连接。之后，开发了一种递归优化算法来推断最大后验概率（MAP）解决方案，该解决方案通过检测、分割和更新三个步骤之间的交替来每次处理一个实例。建议的ERIS框架主要在两个方面不同于以前的工作。首先，它是基于样本的和无模型的，它可以实现特定对象类的实例级分割，只需给出少数（通常少于10个）带注释的样本。这样的优点使得它能够在没有大量手动标记的数据可用于训练强分类模型的情况下使用，这是大多数现有方法所要求的。其次，我们的递归优化策略允许在完整的假设空间中进行合理有效的MAP推理，而不是试图在一个单一的镜头中推断出解决方案，这具有极高的计算复杂性。ERIS框架在这项工作中的具体应用植物叶片分割。在公共基准上进行了实验，以证明我们的方法在有效性和效率方面的优越性，与最先进的。
Instance segmentation is a challenging computer vision problem which lies at the intersection of object detection and semantic segmentation. Motivated by plant image analysis in the context of plant phenotyping, a recently emerging application field of computer vision, this paper presents the exemplar-based recursive instance segmentation (ERIS) framework. A three-layer probabilistic model is first introduced to jointly represent hypotheses, voting elements, instance labels, and their connections. Afterward, a recursive optimization algorithm is developed to infer the maximum a posteriori (MAP) solution, which handles one instance at a time by alternating among the three steps of detection, segmentation, and update. The proposed ERIS framework departs from previous works mainly in two respects. First, it is exemplar-based and model-free, which can achieve instance-level segmentation of a specific object class given only a handful of (typically less than 10) annotated exemplars. Such a merit enables its use in case that no massive manually-labeled data is available for training strong classification models, as required by most existing methods. Second, instead of attempting to infer the solution in a single shot, which suffers from extremely high computational complexity, our recursive optimization strategy allows for reasonably efficient MAP-inference in full hypothesis space. The ERIS framework is substantialized for the specific application of plant leaf segmentation in this work. Experiments are conducted on public benchmarks to demonstrate the superiority of our method in both effectiveness and efficiency in comparison with the state-of-the-art.
登录
查看更多内容
DOI：
10.1109/tcsvt.2015.2397200
发表时间：
2016-04
影响因子：
8.4
作者：
Jianfang Hu;Weishi Zheng;J. Lai;S. Gong;T. Xiang
通讯作者：
Jianfang Hu;Weishi Zheng;J. Lai;S. Gong;T. Xiang
DOI：
10.1109/cvpr.2006.305
发表时间：
2006-06
期刊：
2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)
影响因子：
--
作者：
J. Winn;J. Shotton
通讯作者：
J. Winn;J. Shotton
影响因子：
5.1
作者：
Minervini, Massimo;Abdelsamea, Mohammed M.;Tsaftaris, Sotirios A.
通讯作者：
Tsaftaris, Sotirios A.
影响因子：
3.3
作者：
Scharr, Hanno;Minervini, Massimo;Tsaftaris, Sotirios A.
通讯作者：
Tsaftaris, Sotirios A.
DOI：
10.1109/cvpr.2017.100
发表时间：
2017-04
期刊：
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
影响因子：
--
作者：
Anurag Arnab;Philip H. S. Torr
通讯作者：
Anurag Arnab;Philip H. S. Torr