Yoshinori Konishi

Learn More
This paper presents a high performance JGCF pose estimation system based on the newly-proposed sparse Bayesian regression technique (Relevance Vector Machine , RVM) and sparse representation of facial patterns. In our system, after localizing 20 key facial points, sparse features of these points are extracted to represent facial property, and thenRVM is(More)
This paper proposes a new handwritten character recognition method using a statistical technique based on Bayes' theorem. This technique is the so-called Bayesian filter. The Bayesian filter was originally intended for spam mail recognition, and is known to be a very effective approach. Consequently, we applied the Bayesian filter method to recognition of(More)
Lazy Learning is a new modeling method using memory-based learning. It is applied to control methods using a modeling technique. This paper proposes a new method that can be applied to position control. This method provides a way of setting the query point. Lazy Learning thus performs as a controller in the proposed method, not as an inverse model of the(More)
A fundamental step to solve bin-picking and grasping problems is the accurate estimation of an object 3D pose. Such visual task usually rely on profusely textured objects: standard procedures such as detection of interest points or computation of appearance-based descriptors are favoured by using a highly informative surface. However, texture-less objects(More)
We present a robust and efficient framework for facial shape model fitting. Traditional model fitting approaches are sensitive to noise resulting from scene variations due to lighting, facial expressions, poses, etc., and tend to spend substantial computational effort due to heuris-tic searching algorithms. Our work distinguishes itself from conventional(More)
This paper proposes an effective control system for a container crane that takes into account disturbances such as a wind. For efficient crane operation, it is necessary to realize accurate positioning of cranes, to minimize oscillation of containers and to realize the fastest possible control time while they are being carried. The crane input is assumed to(More)
We propose a novel image feature for textureless object detection. The feature is based on quantized gradient orientations those have been shown to be robust to cluttered backgrounds and illumination changes. We make this feature robust to the appearance changes of a targeted object itself induced by its transformations and small deformations. In our(More)
We present that the application of the Nash bargaining model, which is a solution method in the bargaining game theory, to setting the utility function for integrated design which has some design purposes. The bargaining game theory is often introduced to explain situations in economic activity, but it is also applicable to engineering problem. Applying the(More)
This paper proposes a new hybrid controller for a 2-mass spring system with stick-slip motion. The 2-mass spring system is considered as a hybrid system with two states: a static friction state and a kinetic friction state. In controller design, the hybrid system is described as a piecewise affine (PWA) system, which is one of the modeling frameworks of(More)