摘要:
A method to reduce a neural network includes: adding a reduced layer, which is reduced from a layer in the neural network, to the neural network; computing a layer loss and a result loss with respect to the reduced layer based on the layer and the reduced layer; and determining a parameter of the reduced layer based on the layer loss and the result loss.
摘要:
A high precision signal sensing system and method using an infrared light is provided. The high precision signal sensing system may receive, from a light emitting device, a plurality of lights including a first light and a second light, may measure intensities of the first light and the second light, and may measure a light emitting intensity of the light emitting device based on an intensity difference between the measured light receiving intensities.
摘要:
Provided is a method and apparatus of extracting a 3D facial expression of a user. When a facial image of the user is received, the 3D facial expression extracting method and apparatus may generate 3D expression information by tracking an expression of the user from the facial image using at least one of shape-based tracking and texture-based tracking, may generate a 3D expression model based on the 3D expression information, and reconstruct the 3D expression model to have a natural facial expression by adding muscle control points to the 3D expression model.
摘要:
A three-dimensional (3D) space interface apparatus and method are provided. The 3D space interface apparatus provides an interface in which a user may directly touch and transform a 3D virtual object of a 3D space, which is shown in various ways depending on a position of the user. The 3D space interface apparatus may provide for manipulation of the 3D virtual object by the user using a sensing signal that senses the position and movement of a manipulation object the user. Specifically, an interface unit of the apparatus verifies, using the sensing signal, whether a predetermined movement of the manipulation object has occurred or whether a contact between the 3D virtual object and the manipulation object has been made, and generates a predetermined reaction corresponding to the predetermined movement or the contact.
摘要:
A face tracking apparatus includes: a face region detector; a segmentation unit; an occlusion probability calculator; and a tracking unit. The face region detector is configured to detect a face region based on an input image. The segmentation unit is configured to segment the face region into a plurality of sub-regions. The occlusion probability calculator configured to calculate occlusion probabilities for the plurality of sub-regions. The tracking unit is configured to track a face included in the input image based on the occlusion probabilities.
摘要:
A system, apparatus, and method for precisely estimating a three-dimensional (3D) position and a direction. The 3D position and direction estimation apparatus may estimate a distance between at least one receiver and at least one transmitter and a direction of a remote device, based on intensity information of a signal measured at the at least one receiver, may sequentially select the minimum number of intensity information for estimating the 3D position and the direction of the remote device, in a descending order of robustness against noise, based on the estimated distance and direction of the remote device, and may precisely estimate the 3D position and the direction of the remote device based on the selected intensity information.