Abstract:
An estimator training method and a pose estimating method using a depth image are disclosed, in which the estimator training method may train an estimator configured to estimate a pose of an object, based on an association between synthetic data and real data, and the pose estimating method may estimate the pose of the object using the trained estimator.
Abstract:
An estimator training method and a pose estimating method using a depth image are disclosed, in which the estimator training method may train an estimator configured to estimate a pose of an object, based on an association between synthetic data and real data, and the pose estimating method may estimate the pose of the object using the trained estimator.
Abstract:
Provided a semiconductor device comprises, a plurality of semiconductor patterns spaced in a first direction; a plurality of mold insulating layers between the plurality of semiconductor patterns, a plurality of silicide patterns contacting the plurality of semiconductor patterns; and a plurality of first metal conductive films between the plurality of mold insulating layers and connected to each of the silicide patterns, wherein each of the silicide patterns includes a first sidewall that faces the semiconductor pattern, and a second sidewall which faces the first metal conductive film, the first sidewall of the silicide pattern and the second sidewall of the silicide pattern extends in the first direction, and the first sidewall of the silicide pattern and the second sidewall of the silicide pattern are curved surfaces.
Abstract:
An unmanned image capturing device and a method of operation thereof are provided. The unmanned image capturing device includes a movement device that moves the unmanned image capturing device; a camera; a communication circuit; a location detection circuit that detects a location of the unmanned image capturing device; a processor; and a memory that stores instructions. The processor moves the unmanned image capturing device, captures a plurality of images, detects a plurality of three-dimensional locations, at which the plurality of images are respectively captured, sends at least one of the captured images and information about the respective detected location of the at least one of the captured images, receives information about a change in the at least one of the captured images or information about an image selection, and controls the movement device to move the unmanned image capturing device to a location based on the received information.
Abstract:
An estimator training method and a pose estimating method using a depth image are disclosed, in which the estimator training method may train an estimator configured to estimate a pose of an object, based on an association between synthetic data and real data, and the pose estimating method may estimate the pose of the object using the trained estimator.
Abstract:
Provided a semiconductor device comprises, a plurality of semiconductor patterns spaced in a first direction; a plurality of mold insulating layers between the plurality of semiconductor patterns, a plurality of silicide patterns contacting the plurality of semiconductor patterns; and a plurality of first metal conductive films between the plurality of mold insulating layers and connected to each of the silicide patterns, wherein each of the silicide patterns includes a first sidewall that faces the semiconductor pattern, and a second sidewall which faces the first metal conductive film, the first sidewall of the silicide pattern and the second sidewall of the silicide pattern extends in the first direction, and the first sidewall of the silicide pattern and the second sidewall of the silicide pattern are curved surfaces.
Abstract:
An unmanned aerial vehicle has a main body, a plurality of propeller connection parts extending from at least one side surface of the main body by a specific length, a plurality of propellers respectively connected to ends of the plurality of propeller connection parts, and a plurality of cameras mounted on at least one surface of the main body. A first camera interposed between the plurality of propeller connection parts among the plurality of cameras is disposed spaced from a center point of the main body by a distance of a first size. A first virtual straight line connecting a center point of a first propeller disposed adjacent to the first camera among the plurality of propellers to a center point of the main body has a length of a second size.
Abstract:
An X-ray imaging method and an apparatus having an improved structure configured to reduce pain and discomfort during a procedure. The X-ray imaging apparatus acquires an X-ray image by compressing an object that is irradiated by an X-ray source. An X-ray detector includes an object contact material that contacts the object. X-rays that pass through the object are detected X-rays, an electrical signal is generated based on the X-rays. A paddle is disposed between the X-ray source and the X-ray detector to be vertically adjustable relative to the X-ray source and X-ray detector so as to compress the object placed on the object contact surface. The paddle includes a supporting member, a pressure member coupled to the supporting member, and at least one adjustment member arranged between the supporting member and the pressure member to regulate movement of the pressure member according to the shape of the object.
Abstract:
Disclosed are a body of an unmanned aerial vehicle and an unmanned aerial vehicle including the same. The body includes a housing arranged in a first direction and including at least four housing sidewalls and an opening defined by the housing sidewalls and configured to receive a battery pack, a propeller support extending a specific distance from each housing sidewall in a second direction away from the center of the housing and perpendicular to the first direction and having, on a distal end, a motor and a propeller connected to the motor, and a PCB disposed on at least one of the housing sidewalls and associated with operating the motor and the propeller.