Human detection using depth information by Kinect

Conventional human detection is mostly done in images taken by visible-light cameras. These methods imitate the detection process that human use. They use features based on gradients, such as histograms of oriented gradients (HOG), or extract interest points in the image, such as scale-invariant feature transform (SIFT), etc. In this paper, we present a novel human detection method using depth information taken by the Kinect for Xbox 360. We propose a model based approach, which detects humans using a 2-D head contour model and a 3-D head surface model. We propose a eye detection by open Cv code . We also explore the tracking algorithm based on our detection result.

The detection of persons in an image has been the subject of several studies. Most of these works were done on images taken by cameras in visible light (RGB). In this paper,we are interested in people contours detection on the Kinect3D images. We investigate the application of Gradient approach and optimal filters on depth images. We also use this detection to monitor the person via her gestures. Results show that edge detection of Canny is good for people in both light condition but, the performance of Sobel algorithm was better for the images taken in the dark depths.

The detection and tracking of a hand is an emerging research issue now-a-days to control the devices by hand motion. Conventional hand detection methods use color and shape information from a RGB camera. With the recent advent of the depth camera, some researchers show that they can improve the performance of hand detection by combining the color (or intensity) information with the information from the depth camera. In this paper, we propose a novel method for hand detection using both color and depth information from Microsoft’s Kinect device. The proposed method extract the candidate hand regions from the depth image and select the best candidate based on the color and shape feature of each candidate regions. Then the contour of the selected candidate is determined in the higher resolution RGB image to improve the positional accuracy. For the tracking of the detected hand, we propose the boundary tracking method based on Generalized Hough Transform (GHT). The experimental results show that proposed method can improve the accuracy of hand motion detection over conventional methods.

Proceedings of the Fifth Symposium on Information and Communication Technology - SoICT '14, 2014

This paper investigates on an approach of how to extract and track multiple subjects from a sequence of depth images. The Kinect camera is used to obtain a depth image revealing the depth information of a scene. Our proposed system includes the object clustering module to segment different isolated regions correspondent to objects in a depth image and foreground detection module to find moving regions from a sequence of frames. The combination of the two modules let us know which object is moving within a sequence of frames to locate a human subject. In order to extract the depth silhouettes of multiple subjects during time, we propose the use of matching algorithm between two consecutive frames to track their movements. We evaluate the algorithm with a long sequence of frames within a complex environment containing backgrounds with furniture and show how the algorithm is able to precisely extract and separate different human subjects with a fast processing speed. Therefore, the proposed approach is suitable for widely practical applications working with human activity recognition, human pose estimation and human tracking from depth images.

In this paper we present a module for detection and segmentation of human torsos for Human-Robot Interaction. People detection is a very important task in Human-Robot Interaction and particularly at the beginning of this interaction. Visual and 3D data coming from a Kinect sensor is used to achieve the task. The module proposed uses three main process: a) a face detector b) a skin color detector t and c) a 3D region growing segmentation technique. Open sources libraries like OpenNI, OpenKinect and Point Cloud Library are used in our implementation to handle the device and sensors data, as well as OpenCV and GLUT to the image processing and display.

The paper addresses the problem of people extraction in a closed context in a video sequence including colour and depth information. The study is based on low cost depth captor included in products such as Kinect or Asus devices that contain a couple of cameras, colour and depth cameras. Depth cameras lack precision especially where a discontinuity in depth occur and some times fail to give an answer. Colour information may be ambiguous to discriminate between background and foreground. This made us use first depth information to achieve a coarse segmentation that is improved with colour information. Furthermore, color information is only used when a classification in two classes of fore/background pixels is clear enough. The developed method provides a reliable and robust segmentation and a natural visual rendering, while maintaining a real time processing.

In this paper, we have proposed a system to keep track of human body movements in real time mode. The Kinect sensors are used to capture Depth and Audio streams. The system is designed by integration of two modules namely Kinect Module and Augmented Reality module. The kinect module performs Voice Recognition and captures depth images that are used by Augmented Reality module for computing the distance parameters. Augmented Reality module also captures real-time image data streams from high resolution camera. The system generates 3D module that is superimposed on real time data.

Object segmentation is a fundamental research topic in computer vision. While, only the color information for object segmentation has been the main focus of research, with the availability of low cost color plus range sensors, depth segmentation is now attracting significant attention. This paper presents a novel algorithm for depth segmentation. The proposed technique exploits the divergence of the 2D vector field to segment three-dimensional (3D) object in the depth maps. For a given depth image acquired using a low resolution Kinect sensor, a 2D vector field is computed first at each point of the range image. The depth map is then converted to the div map by computing the 2D vector field’s divergence. The latter maps the vector field to a scalar field. The variation of divergence values over the surface contour of the 3D object helps to extract its boundaries. Finally, the depth segmentation is accomplished by applying a threshold to the div map to segment 3D object from the background. In addition to removing the background, the proposed technique also segments the object from the surface on which the object is positioned. The proposed technique was tested on low resolution Washington RGB-D (Kinect) object dataset. Preliminary experimental results suggest that the proposed algorithm achieves better depth segmentation compared to state-of-the art graph-based depth segmentation. The proposed technique also outperforms the latter by achieving 40% higher computational efficiency.

Multimedia Tools and Applications, 2019

In this paper we present a new 3D descriptor for human classification and a human detection method based on this descriptor. The proposed 3D descriptor allows for the classification of an object represented by a point cloud, as human or non-human. It is derived from the well-known Histogram of Oriented Gradient by employing surface normals instead of gradients. The process consists in an appropriate subdivision of the object point cloud into blocks. These blocks provide the spatial distribution modeling of the surface normal orientation into the different parts of the object. This distribution modelling is expressed in the form of a histogram. In addition we have set up a multi-kinect acquisition system that provides us with Complete Point Clouds (CPC) (i.e. 360 • view). Such CPCs enable a suitable processing, particularly in case of occlusions. Moreover they allow for the determination of the human frontal orientation. Based on the proposed 3D descriptor, we have developed a human detection method that is applied on CPCs. First, we evaluated the 3D descriptor over a set of CPC candidates by using the Support Vector Machine (SVM) classifier. The learning process was conducted with the original CPC database that we have built. The results are very promising. The descriptor can discriminate human from non-human candidates and provides the frontal direction of the humans with high precision. In addition we demonstrated that using the CPCs improves significantly the classification results in comparison with Single Point Clouds

Hand detection is one of the most important and crucial step in human computer interaction environment. This paper presents a distance technique for hand detection based on depth image information get from Kinect sensor. Distance was used as the method of this study for hand detection. First, Kinect sensor was used to obtain depth image information. Second, background subtraction and iterative method for shadow removal applied to reduce noise from the depth image. Then, it used the official Microsoft SDK for extraction process. Finally, two hands could be segmented based on different color in specific distance. The experiment result shows that hands and head can be detected in a different position with good accuracy.