Electronics and Telecommunications Research Institute

Ho Wook Jang

Jung Jae Yu

Seung Woo Nam

Hae Dong Kim

Myung Ha Kim

Kyung Ho Jang

Yun Ji Ban

Do Hyung Kim

Seong Jin Park

Seong Jun Bae

Jae Woo Kim

Hye-Sun Kim

Ji Young Park

Myung Joon Kim

Jin Seo Kim

Kyung Ae Moon

Hyuk Jeong

Jin Ho Kim

Wook Ho Son

Kwang Hyun Shim

Moon Wook Ryu

Hye Sun Kim

Hee Kwon Kim

Soon Chan Park

Ju Yong Chang

Young Ha Hwang

Disclosed is an apparatus and method of segmenting an object. An object segmentation method according to the present disclosure includes: receiving an input image; receiving a user input indicating at least one piece of information on a foreground region and a background region included in the input image; generating at least one among a foreground pixel list and a background pixel list using the received user input; calculating Gaussian distribution of at least one pixel that makes up the input image using at least one among the generated foreground pixel list and background pixel list; and determining whether the at least one pixel is a foreground pixel or a background pixel using the calculated Gaussian distribution.

Object segmentation apparatus and method using Gaussian mixture model and total variation

An arbitrary viewpoint image generation method includes obtaining an original image set photographed by a plurality of cameras included in a camera array at each of at least one focal distance at a same time point; obtaining a multi-focus image set by generating a multi-focus image from the original image set for each of the at least one focal distance; and generating an arbitrary viewpoint image at a position where a viewpoint is to be moved from the multi-focus image set.

Apparatus and method for generating image of arbitrary viewpoint using camera array and multi-focus image

A method of increasing a photographing speed of a photographing device which capture an image through a combination of two or more photographing devices and generate and provide an image by using the captured image, thereby increasing a photographing speed. An RGB image obtaining device and a depth image obtaining device alternately perform photographing to obtain an image. Also, a second depth image and a second RGB image respectively corresponding to a first RGB image and a first depth image which are alternately obtained by performing alternate photographing are synthesized and output, thereby actually increasing a photographing speed by twice.

Method of increasing photographing speed of photographing device

Disclosed are a system for making 3D contents provided with visual fatigue minimization and a method of the same. More particularly, an exemplary embodiment of the present invention provides a system for making 3D contents including: a human factor information unit generating guide information for making 3D contents by considering factors causing visual fatigue of the 3D contents; and a 3D contents making unit applying guide information generated by the human factor information unit to 3D contents data inputted for making the 3D contents to make the 3D contents, and a method of making 3D contents.

System for making 3D contents provided with visual fatigue minimization and method of the same

A method of converting a two-dimensional video to a three-dimensional video, the method comprising: comparing an image of an nframe with an accumulated image until an n&#x2212;1frame in the two-dimensional video to calculate a difference in a color value for each pixel; generating a difference image including information on a change in a color value for each pixel of the nframe; storing an accumulated image until the nframe by accumulating the information on the change in the color value for each pixel until the nframe; performing an operation for a pixel in which a change in a color value is equal to or larger than a predetermined level by using the difference image to generate a division image and a depth map image; and converting the image of the nframe to a three-dimensional image by using the depth map image.

Method and apparatus for converting 2D video image into 3D video image

Disclosed are an apparatus and a method for extracting a foreground layer from an image sequence that extract a foreground object layer area in which a depth value is discontinuous with that of a background from an input image sequence. By using the present disclosure, the layer area is automatically tracked in the subsequent frames through user's setting in the start frame in the image sequence in which the depth values of the foreground and the background are discontinuous, thereby extracting the foreground layer area in which the drift phenomenon and the flickering phenomenon are reduced.

Apparatus and method for extracting foreground layer in image sequence

An apparatus for correcting a stereoscopic image using matching information, includes: a matching information visualizer receiving input of original stereoscopic images and intuitive matching information and visualizing a pair of stereoscopic images based on the intuitive matching information; a correction information processor obtaining a statistical camera parameter based on the intuitive matching information and correcting the received stereoscopic image using the statistical camera parameter; and an error allowable controller providing allowable error information to the correction information processor in consideration of an error allowable degree according to a selected time from the received intuitive matching information and preset human factor guide information, to extract a correlation between stereoscopic images using a stereoscopic image and provided information, thereby helping such that an erroneously photographed image is correctly photographed or correcting the image such that the erroneously photographed image is correctly interpreted, which leads to minimization of visual fatigue.

Apparatus and method for correcting stereoscopic image using matching information

Disclosed is a method for embedding and extracting a spatial domain blind watermark using a sample expansion. The sample expansion method using a sample expansion coefficient is introduced to basically improve a common disadvantage of the conventional watermarking method based on the patchwork method, for instance, the watermark is not robust against a compression attack such as JPEG. By applying the sample expansion to a sample group, the embedded watermark comes to contain a low frequency component more larger than ever, so that radio frequency component decreases as much. Thus, although the radio frequency component of the watermark is damaged by the image compression such as JPEG, it is possible to relatively reduce the damage of the watermark. Also, an adaptive watermark embedding method can be considered such that the embedded watermark value is proportional to an average value of the pixel. In order to overcome non-robustness that is a common disadvantage of the conventional methods based on the conventional patchwork method or the conventional modified patchwork method, not a single pixel but sample-expanded pixel is chosen, and there is proposed a spatial domain blind watermarking method using a watermark value that is proportional to an average value of pixel every tile.

Method for embedding and extracting a spatial domain blind watermark using sample expansion

Growing community of inventors

Daejeon, South Korea

Ho Wook Jang