3i Corporation

31, Incorporated

Ji Wuck Jung

Ken Kim

Farkhod Rustam Ugli Khudayberganov

Kevin Nathanael Santoso

Mikhail Li

Dong Hee Lee

Byung Ju Lee

Donghee Lee

Farkhod Khudayberganov

Hun Joo Song

Ji Soo Kang

Seo Jae Kweon

Fernando Carrasco Ferretti

A hole-filling method, performable on a computing device, for providing a three-dimensional virtual model according to a technical aspect of the present application may comprises the operations of: acquiring an original training image and a hole creation training image, wherein the hole creation training image is an image in which at least one hole is created based on the original training image; creating a hole-filling training image by performing hole-filling on the hole creation training image using a neural network; performing spherical transformation on each of the hole-filling training image and the original training images; and training the neural network based on the difference between the spherically transformed hole-filling training image and the spherically transformed original training image.

Pre-processing method for creating 3D virtual model and computing device therefor

A method of tracking an object performed in a portable terminal linked with a rotatable cradle so that a rotation of the rotatable cradle is controlled to track the object, and the method includes receiving frame images captured in a forward direction at a preset frame rate, and detecting whether an object exists in the frame image for each of the frame images, when the object is detected in the frame image for each of the frame images, calculating a reference distance in association with a position of the detected object in the frame image, calculating a rotational speed of the rotatable cradle for a first frame image based on accumulation of reference distances of a plurality of frame images included in a frame group including the first frame image, and controlling driving of the rotatable cradle based on the rotational speed.

Method for tracking object by controlling a rotation of a cradle using a portable terminal and system thereof

In order to calculate an actual distance between coordinates in an image, a regular image for an original image may be generated based on an intrinsic parameter of a camera and pose information of the camera, based on first coordinates in the regular image that are received, first world coordinates corresponding to the first coordinates may be calculated in a world coordinate system, based on second coordinates in the regular image that are received, second world coordinates corresponding to the second coordinates may be calculated in the world coordinate system, and a distance between the first world coordinates and the second coordinates may be calculated.

Method and apparatus for calculating actual distance between coordinates in image

A computing device is provided. The computing device includes a memory configured to store one or more instructions; and a processor configured to execute the one or more instructions stored in the memory, in which the processor may execute the one or more instructions to generate a first predicted depth map image for a learning red/green/blue (RGB) image using a neural network, and train the neural network based on a difference between a learning depth map image having depth information and the first predicted depth map image. The learning depth map image may match 1:1 and may be generated based on the same setting change for a basic spherical image.

Depth map image generation method and computing device therefor

A server is provided. The server provides virtual indoor space content corresponding to a real indoor space. The server includes a memory; a communication module; and at least one processor operatively connected to the memory and the communication module. The at least one processor may receive image data to generate the virtual indoor space content from a mobile terminal through the communication module, the image data including, for each of a plurality of indoor points, an omnidirectional image acquired from each indoor point and relative location information on the indoor point, and generate the virtual indoor space content by mapping a plurality of virtual points corresponding to the plurality of indoor points in a virtual indoor space based on the relative location information.

Method of providing virtual indoor space content and server therefor

Stand for portable terminal

In order to output an image to a user, an electronic device may include loading a plurality of basic image tiles with a first resolution, the plurality of basic image tiles being preset for a graphic space set based on a virtual coordinate system, determining a target basic-image tile among the plurality of basic image tiles based on a coordinate of a virtual camera set on the virtual coordinate system, transmitting information on a target basic-image tile to a server, receiving a plurality of detailed image tiles with a second resolution corresponding to the target basic-image tile from the server, and outputting an image with the second resolution based on the plurality of detailed image tiles when the virtual camera zooms in on the target basic-image tile.

Method and device for outputting an image

A mobile device which operates in conjunction with a mobile device cradle is provided. The mobile device is a mobile device which is held by a mobile device cradle and controls the mobile device cradle to be driven, and may include at least one camera which captures a forward image, a display, and at least one processor which executes an application which detects a subject from the forward image and controls the mobile device cradle to be driven so that a forward direction of the mobile device follows the subject according to movement of the subject.

Mobile device and method of controlling mobile device cradle

Provided is a method performed in a portable terminal mounted on a rotatable terminal cradle to generate a 360° captured image by controlling operations of the terminal cradle. The 360° captured image is used generate a virtual space corresponding to a captured real space. Communication is established with the terminal cradle; an image generation input is received from a user; a first captured image is generated by imaging in an imaging direction that the terminal faces; driving commands are transmitted to the terminal cradle to rotate the imaging direction of the terminal clockwise or counterclockwise; a second captured image is generated by imaging in the changed imaging direction when the terminal cradle rotates according to the driving command; and presence of a human object is determined by recognizing objects in the first and second captured images, and re-imaging is performed in the imaging direction when a human object is present.

Image generation method using terminal cradle and portable terminal therefor

A hole filling method for providing a virtual three-dimensional (3D) model, the hole filling method being executable on a computing device and including: acquiring an original learning image and a hole generation learning image for the original learning image, wherein the hole generation learning image is an image in which at least one hole is generated based on the original learning image; generating a hole filling learning image by performing hole filling on the hole generation learning image by using a neural network; performing a spherical transformation on each of the hole filling learning image and the original learning image; and training the neural network based on a difference between the spherically-transformed hole filling learning image and the spherically-transformed original learning image.

Hole filling method for providing virtual three-dimensional model and computing device therefor

The described technology is a technique related to an indoor localization method using deep learning. The indoor localization method comprises: opening a 3D tour comprising a plurality of panoramic images; receiving a first perspective image captured by a camera provided in the user device; calculating global features for the first perspective image and each of the plurality of panoramic images included in the 3D tour; selecting a most similar panoramic image to the first perspective image by using the calculated global features; computing an indoor location corresponding to a location of the camera on the 3D tour by using feature points included in the selected panoramic image and the first perspective image; and providing the computed indoor location to the user device.

Method for indoor localization using deep learning

The described technology is a technique related to a 3D tour photographing apparatus and method. The 3D tour photographing apparatus comprising: a camera module for photographing a plurality of images for generating an around-view image for a particular point, a location tracking module for detecting a location change or direction change of the camera module, a tilt unit for changing a position or angle between the camera module and the location tracking module, and a control module for deriving a movement path based on data provided from the location tracking module, and for mapping the plurality of images provided from the camera module onto the movement path.

3D tour photographing apparatus and method

The described technology is a technique related to a server and method for displaying additional information on a 3D tour. In this case, the method for displaying additional information on a 3D tour performed by a 3D tour management server linked to a user device, the method comprising: providing the user device with a floor plan for a particular space and an interface capable of inserting additional information based on the floor plan, receiving a plurality photographed images for configuring a 3D tour for the particular space from the user device, in a process of photographing the 3D tour performed by the user device, receiving additional information as input from the user device via the interface, combining a 3D tour configured based on the plurality photographed images and the additional information inputted via the interface, and displaying the additional information onto the 3D tour or the floor plan.

Server and method for displaying additional information on 3D tour

Provided is a device cradle. The device cradle includes a fixer for fixing a device, a body portion including a main body and a penetrating groove which is formed to penetrate the main body from a top surface of the main body and accommodates at least a portion of the fixer, and a first driving portion for rotating the fixer by a specified angle around a horizontal axis with respect to the body portion, wherein the penetrating groove has a spherical shape, and wherein at least part of the fixer may be formed in a spherical shape.

Device cradle

The described technology is a technique related to a system and method for displaying 3D tour comparisons. The method for displaying a 3D tour comparison including setting each reference point for first and second tours that are different from each other, and setting coordinate values for a plurality of image photographing points included in the first and second tours, respectively, in the first tour, setting a first point that is one of the plurality of image photographing points included in the first tour, in the second tour, selecting a second point closest to the coordinate value of the first point out of the plurality of image photographing points included in the second tour, and displaying an image related to the first point on a part of a split-screen of a split-screen, and displaying an image related to the second point on at least a part of the rest of the split-screen of the split-screen.

Growing community of inventors

Goyang-si, South Korea

Ji Wuck Jung