Kinect 下深度信息获取技术及其在三维目标识别中的应用综述

微软公司 2010 年推出的 Kinect 深度传感器能够同步提供场景深度和彩色信息，其应用的一个关键领域就是目标 识别。传统的目标识别大多限制在特殊的情形，如：手势识别、人脸识别，而大规模的目标识别是近年来的研究趋势。 通过 Kinect 得到的 RGB-D 数据集多为室内和办公环境下获取的多场景、多视角、分目标类型的数据集，为大规模的目标 识别算法设计提供了学习基础。同时，Kinect 获取的深度信息为目标识别提供了强有力的线索，利用深度信息的识别方法 较以前的方法具有无法比拟的优势，大大地提高了识别的精度。文章首先对 Kinect 的深度获取技术做了详细介绍；其次 对现有的 3D 目标识别方法进行综述，接着对已有的 3D 测试数据集进行分析和比较；最后对文章进行小结以及对未来 3D 目标识别算法和 3D 测试数据集的发展趋势作了简单的阐述。

Review on the Technique to Obtain Depth Information Using Kinect and Its Application to Three Dimensional Object Recognition

In 2010, Microsoft launched a depth camera named Kinect, which can synchronously provide the depth and RGB information of a scene. Using sensing technique, robust depth information can be achieved at real time. One key area of Kinect is object recognition. Most of the traditional object recognition techniques are limited in special situations, such as gesture recognition, face recognition. Recent trend in computer vision is large-scale recognition. We can get the RGB-D datasets from various scenes, categories, instances and viewpoints of real domestic and office environment by Kinect. The RGB-D datasets can meet the need of practical applications. The depth information provides a strong clue for object recognition. Compared with the previous methods, using depth information for object recognition has incomparable advantage, which can greatly improve the accuracy of recognition. The four major components of this paper are as follows: 1) a detailed introduction of the technique to obtain depth information using Kinect; 2) review of the existing 3D Object recognition methods; 3) an analysis and comparison of the existing 3D Datasets for testing; 4) summary of the article, and the future development trend of the 3D object recognition algorithms and the 3D datasets for testing.