Daisaku Arita

Dynamic Control of Adaptive Mixture-of-Gaussians Background Model

We propose a method for create a background model in non-stationary scenes. Each pixel has a dynamic Gaussian mixture model. Our approach can automatically change the number of Gaussians in each pixel. The number of Gaussians increases when pixel values often change because of Illumination change, object moving and so on. On the other hand, when pixel values are constant in a while, some Gaussians are eliminated or integrated. This process helps reduce computational time. We conducted experiments to investigate the effectiveness of our approach.

Recognition of local features for camera-based sign language recognition system

A sign language recognition system is required to use information from both global features, such as hand movement and location, and local features, such as hand shape and orientation. We present an adequate local feature recognizer for a sign language recognition system. Our basic approach is to represent the hand images extracted from sign-language images as symbols which correspond to clusters by a clustering technique. The clusters are created from a training set of extracted hand images so that a similar appearance can be classified into the same cluster on an eigenspace. The experimental results indicate that our system can recognize a sign language word even in two-handed and hand-to-hand contact cases.

A fast algorithm for adaptive background model construction using parzen density estimation

Non-parametric representation of pixel intensity distribution is quite effective to construct proper background model and to detect foreground objects accurately. However, from the viewpoint of practical application, the computation cost of the distribution estimation should be reduced. In this paper, we present fast estimation of the probability density function (PDF) of pixel value using Parzen density estimation and foreground object detection based on the estimated PDF. Here, the PDF is computed by partially updating the PDF estimated at the previous frame, and it greatly reduces the computation cost of the PDF estimation. Thus, the background model adapts quickly to changes in the scene and, therefore, foreground objects can be robustly detected. Several experiments show the effectiveness of our approach.

A real-time motion capture system with multiple camera fusion

This paper presents a real-time motion capture system of 3D multi-part objects, whose purpose is to do seamless mapping of objects in the real world into virtual environments easily. In general, virtual environment applications such as man-machine seamless interaction require the system to estimate accurate motion parameters at real-time for natural objects such as human bodies. To achieve this requirement, we have been developing a vision-based motion capture system which reconstructs time-varying motion parameters of 3D multi-part objects. The advantage of such a vision-based system is that it is possible to acquire the other scene parameters such as shape and surface properties at the same time, using the same equipment in measuring motion. In this paper, as our first system, we have implemented a color-marker-based motion capture system which realizes multi-view fusion and have demonstrated our motion capture and reconstruction system works at real-time on PC-clusters.

RPV-II: A Stream-Based Real-Time Parallel Vision System and Its Application to Real-Time Volume Reconstruction

In this paper, we present RPV-II, a stream-based real-time parallel image processing environment on distributed parallel computers, or PC-cluster, and its performance evaluation using a realistic application. The system is based on our previous PC-cluster system for real-time image processing and computer vision, and is designed to overcome the problems of our previous system, one of which is long latency when we use pipelined structures. This becomes a serious problem when we apply the system to interactive applications. To make the latency shorter, we have introduced stream data transfer, or fine grained data transfer, to RPV-II. One frame data is divided into small elements such as pixels, lines and voxels, and we have developed efficient real-time data transfer mechanism of those. Using RPV-II we have developed a real-time volume reconstruction system by visual volume intersection method, and we have measured the system performance. Experimental results show better performance than that of our previous system, RPV.

Towards robust object detection: integrated background modeling based on spatio-temporal features

We propose a sophisticated method for background modeling based on spatio-temporal features. It consists of three complementary approaches: pixel-level background modeling, region-level one and frame-level one. The pixel-level background model uses the probability density function to approximate background model. The PDF is estimated non-parametrically by using Parzen density estimation. The region-level model is based on the evaluation of the local texture around each pixel while reducing the effects of variations in lighting. The frame-level model detects sudden, global changes of the the image brightness and estimates a present background image from input image referring to a background model image. Then, objects are extracted by background subtraction. Fusing their approaches realizes robust object detection under varying illumination, which is shown in several experiments.

Real-Time 3d hand shape estimation based on inverse kinematics and physical constraints

We are researching for real-time hand shape estimation, which we are going to apply to user interface and interactive applications. We have employed a computer vision approach, since unwired sensing provides restriction-free observation, or a natural way of sensing. The problem is that since a human hand has many joints, it has geometrically high degrees of freedom, which makes hand shape estimation difficult. For example, we have to deal with a self-occlusion problem and a large amount of computation. At the same time, a human hand has several physical constraints, i.e., each joint has a movable range and interdependence, which can potentially reduce the search space of hand shape estimation. This paper proposes a novel method to estimate 3D hand shapes in real-time by using shape features acquired from camera images and physical hand constraints heuristically introduced. We have made preliminary experiments using multiple cameras under uncomplicated background. We show experimental results in order to verify the effectiveness of our proposed method.

Object Detection under Varying Illumination Based on Adaptive Background Modeling Considering Spatial Locality

We propose a new method for background modeling. Our method is based on the two complementary approaches. One uses the probability density function(PDF) to approximate background model. The PDF is estimated non-parametrically by using Parzen density estimation. And foreground object is detected based on the estimated PDF. The other method is based on the evaluation of the local texture at pixel-level resolution while reducing the effects of variations in lighting. Fusing their approach realize robust object detection under varying illumination. Several experiments show the effectiveness of our approach.

Real-time human motion sensing based on vision-based inverse kinematics for interactive applications

Vision-based human motion sensing has a strong merit that it does not impose any physical restrictions on humans, which provides a natural way of measuring human motion. However, its real-time processing is not easy to realize, because a human body has a high degrees of freedom, whose vision-based analysis is not simple and is usually time consuming. Here, we have developed a method in which human postures are analyzed from a limited number of visual cues. It is a combination of numerical analysis of inverse kinematics and visual search. Our method is based on a general framework of inverse kinematics, and, therefore, we can use relatively complex human figure model, which can generate natural human motion. In our experimental studies, we show that our implemented system works in real-time on a PC-cluster.

Real-time computer vision on PC-cluster and its application to real-time motion capture

In this paper, we describe a PC cluster system for real-time computer vision. For easy construction of real-time distributed computer vision on PC cluster, we have developed a programming environment, in which a programmer have to describe only data flow between PCs and processing algorithms on each PC. And we also describe a real-time human motion capture system using multiple cameras as a prototypical application on the PC cluster system, which shows that the system works in real-time.