Electronics and Telecommunications Research Institute

Ho Chul Shin

Eul Gyoon Lim

Ji Ho Chang

Dae Hwan Hwang

Jae Il Cho

Seung Min Choi

Hye Jin Kim

Jae Yeon Lee

Jae Chan Jeong

Eui Gyoon Lim

Kuk Jin Yoon

Dan Hwan Hwang

Jae Chan Jung

There are provided a stereo matching method capable of increasing a resolution and precision of an object and a device for performing the method. The method includes investigating a pattern according to a predetermined period; obtaining left and right stereo images in which the investigated pattern is included or not included; determining whether each of the stereo image obtained according to the predetermined period and a buffering image that is a previous frame image of the stereo image includes the pattern; and generating a final cost volume by performing different image processing according to whether each of the stereo image and the buffering image includes the pattern. Therefore, it is possible to increase precision of a disparity of a thin object and increase accuracy and precision of an object at a long distance.

Stereo matching method and device for performing the method

Disclosed is a method of correcting an input image when projecting the input image using an image projection device. The method includes obtaining information on a projection surface on which the input image is to be projected and information on a user's position through at least one image sensor, performing geometric correction on the input image based on the information on the projection surface and the information on the user's position, performing color correction on the input image based on the information on the projection surface, and performing brightness correction on the input image based on the information on the projection surface.

Apparatus and method for correcting image for image projection device with cognitive function on user and environment

An image matching apparatus includes a bilateral filter that filters a left image and a right image to output a second left image and a second right image; a census cost calculation unit performing census transform on a window based on a first pixel of the second left image and a window based on a second pixel of the second right image to calculate a census cost corresponding to a pair of pixels of the first and second pixels; a support weight calculation unit obtaining support weights of the left and right images or the second left and second right images; a cost aggregation unit obtaining energy values of nodes corresponding to the pair of pixels of the first and second pixels using the census cost and the support weights; and a dynamic programming unit performing image matching using dynamic programming by the energy values of each node obtained.

Stereo matching by census transform and support weight cost aggregation

Disclosed are projector image correction device and method using a range image camera in order to efficiently use a beam projector in an arbitrary environment. In the projector image correction device, range information of a projected surface is determined by using a range image camera to detect a surface shape of the projected surface and determine a surface state such as a color and a material of the projected surface and a state of a surrounding illumination, thereby efficiently correcting an image to be outputted to perform an efficient image output in an arbitrary environment having various colors and materials.

Projector image correction device and method

Provided are an apparatus and method which detect a user's upper body posture and the user's hand postures through at least two cameras to configure a user interface such as computers and home appliances. An apparatus for detecting upper body posture and hand posture includes a stereo vision calculation unit, a color pattern categorization unit, a plurality of position detection units, an upper body posture recognition unit, and a hand posture recognition unit. The stereo vision calculation unit calculates three-dimensional distance information from each image provided from at least two cameras. The color pattern categorization unit categorizes each color and pattern from the image to output color information and pattern information. The position detection units extract three-dimensional position information for each subject based on the three-dimensional distance information, the color information and the pattern information. The upper body posture recognition unit recognizes an upper body posture based on the three-dimensional position information. The hand posture recognition unit recognizes hand postures with hand position information of the three-dimensional position information.

Apparatus and method for detecting upper body posture and hand posture

Provided is a device for improving stereo matching results. The device for improving stereo matching results includes: a stereo camera unit outputting binocular disparity images by using binocular disparity between two images preprocessed according to a plurality of preprocessing conditions; a discrete cosine transform (DCT) unit generating DCT coefficients by performing DCT on the binocular disparity images; a streak estimation unit receiving the DCT coefficients and estimating amounts of streaks distributed on a screen by using AC coefficients, including streak patterns, of the DCT coefficients; a condition estimation unit estimating a preprocessing condition, corresponding to the smallest amount of streaks of the estimated amounts of streaks, of the plurality of preprocessing conditions, as an optimal condition, and a streak removal unit generating binocular disparity images without the streaks by changing predetermined AC coefficients of the DCT coefficients and performing inverse DCT on the changed DCT coefficients.

Device for improving stereo matching results, method of improving stereo matching results using the device, and system for receiving stereo matching results

Provided are a human recognition apparatus and a human recognition method identifying a user based on a walking pattern. The human recognition apparatus includes a detecting unit detecting a vibration according to a user's walking, and outputting an electric signal, a pattern calculating unit acquiring a walker's walking pattern from the electric signal, and a user determining unit comparing the walking pattern with a previously measured reference data by user and identifying the user based on the comparison result. The human recognition apparatus and the human recognition method are robust against peripheral noise and can increase an acceptance rate through a simple structure and procedure by using the waling pattern, which is one-dimensional time information requiring no vast data throughput, as the user identification data.

Growing community of inventors

Daejeon, South Korea

Ho Chul Shin