Semi-supervised learning and feature evaluation for RGB-D object recognition
Semi-supervised learning and feature evaluation for RGB-D object recognition
复制标题
RGB-D 物体识别的半监督学习和特征评估
DOI：
10.1016/j.cviu.2015.05.007
复制
发表时间：
2015-10-01
影响因子：
4.5
通讯作者：
Tan, Tieniu
中科院分区：
文献类型：
--
作者：
Cheng, Yanhua;Zhao, Xin;Tan, Tieniu
随着Kinect等新的深度传感技术提供高质量的同步RGB和深度图像（RGB-D数据），将两种不同的视图结合起来进行对象识别已经引起了计算机视觉和机器人社区的极大兴趣。最近的方法大多采用监督学习方法，这种新的RGB-D模式的基础上的两个特征集。然而，监督学习方法总是依赖于大量的人工标记的数据来训练模型。为了解决这个问题，本文提出了一种半监督学习方法，以减少对大型标注训练集的依赖。该方法可以有效地从相对丰富的未标记数据中学习，如果可以提取RGB和深度视图的强大特征表示。因此，本文提出了一种新的和有效的特征，称为CNN-SPM-RNN，并在统一的半监督学习框架下评估和比较了四个代表性特征（KDES [1]，CKM [2]，HMP [3]和CNN-RNN [4]）。最后，我们验证了我们的方法在三个流行的和公开的RGB-D对象数据库。实验结果表明，该方法仅用20%的标记训练集，就可以在大多数数据库上获得与现有技术相比具有竞争力的性能。(C)2015 Elsevier Inc. All rights reserved.
With new depth sensing technology such as Kinect providing high quality synchronized RGB and depth images (RGB-D data), combining the two distinct views for object recognition has attracted great interest in computer vision and robotics community. Recent methods mostly employ supervised learning methods for this new RGB-D modality based on the two feature sets. However, supervised learning methods always depend on large amount of manually labeled data for training models. To address the problem, this paper proposes a semi-supervised learning method to reduce the dependence on large annotated training sets. The method can effectively learn from relatively plentiful unlabeled data, if powerful feature representations for both the RGB and depth view can be extracted. Thus, a novel and effective feature termed CNN-SPM-RNN is proposed in this paper, and four representative features (KDES [1], CKM [2], HMP [3] and CNN-RNN [4]) are evaluated and compared with ours under the unified semi-supervised learning framework. Finally, we verify our method on three popular and publicly available RGB-D object databases. The experimental results demonstrate that, with only 20% labeled training set, the proposed method can achieve competitive performance compared with the state of the arts on most of the databases. (C) 2015 Elsevier Inc. All rights reserved.