Many classic and contemporary face recognition algorithms work well on public data sets, but degrade sharply when they are used in a real recognition system. This is mostly due to the difficulty of simultaneously handling variations in illumination, image misalignment, and occlusion in the test image. We consider a scenario where the training images are well controlled, and test images are only...

Relationships of 'perceptual distortion' and 'cognitive evaluation' components of body image disturbances to brain activity were investigated. Changes in regional cerebral blood (rCBF) of nine patients with anorexia nervosa restrictive type (AN), 13 patients with bulimia nervosa purging type (BN) and 12 controls following three experiments with single photon emission computed tomography (SPECT)...

OBJECTIVE The authors' goal was to determine whether FDG uptakes in various regions of the brain are different for early and late scanning time in positron emission tomography (PET). METHOD F-18 fluorodeoxyglucose (FDG) PET was performed on 15 healthy normal subjects to obtain early and late acquisition glucose metabolic images (30 and 60 min after FDG injection), respectively. The two sets o...

For user convenience, document image processing captured with a digital camera instead of a scanner has been researched. However, existing methods of document image processing are not usable for a perspective document image captured by a digital camera because most of them are designed for the one captured by a scanner. Thus, we have to rectify the perspective of the document image and obtain t...

In this paper, an original hybrid 2D-3D face recognition approach is proposed using two orthogonal face images, frontal and side views of the face, to reconstruct the complete 3D geometry of the face. This is obtained using a model based solution, in which a 3D template face model is morphed according to the correspondence of a limited set of control points identified on the frontal and side im...

In a face, there is much information of person’s identity. Because this property, various tasks such as expression recognition, identity recognition and deepfake have been actively conducted. Most them use the exact frontal view given face. However, directions face can be observed rather than image in real situation. The profile (side view) lacks when comparing with image. Therefore, if we gene...

Many classic and contemporary face recognition algorithms work well on public data sets, but degrade sharply when they are used in a real recognition system. This is mostly due to the difficulty of simultaneously handling variations in illumination, image misalignment, and occlusion in the test image. We consider a scenario where the training images are well controlled and test images are only ...

Object recognition under varying pose is a challenging problem, especially when illumination variations are also present. In this paper, we propose a pose-normalized view synthesis method under varying illuminations for symmetric objects. For a given non-frontal view of a symmetric object under non-frontal illumination, the mirror image of the original view is equivalent to the view when the ob...

In this paper, we suggest a viewpoint-invariant face recognition model based on view-based representation. The suggested model has four stages: view-based representation, viewpoint classification, frontal face estimation and face recognition. For view-based representation, we obtained the feature space by using independent subspace analysis, the bases of which are grouped like the neurons in th...

Previous studies of visually responsive neurons in the frontal eye fields have identified a selection process preceding saccades during visual search. The goal of this experiment was to determine whether the selection process corresponds to the selection of a conspicuous stimulus or to preparation of the next saccade. This was accomplished with the use of a novel task, called search-step, in wh...