Abstract: A method for generating a depth map includes acquiring at least one image of a scene including a reflective object or a semi-transparent object, acquiring, from the at least one image, a first depth map including a depth value for at least one opaque object which is reflected by the reflective object or viewed through the semi-transparent object, acquiring, from the at least one image, a second depth map including a depth value for the reflective object or the semi-transparent object, and generating a depth map for the scene based on the acquired first depth map and the acquired second depth map, and the generated depth map includes the depth value for the reflective object or the semi-transparent object and the depth value for the at least one opaque object which is reflected by the reflective object or viewed through the semi-transparent object.

Abstract: A three-dimensional depth measuring method using a structured light camera comprising a first projector, a second projector, and an optical sensor, according to the present disclosure, comprises: a step in which a first light pattern is projected by the first projector; a step in which a second light pattern is projected by the second projector; a step of filtering, by a color filter, a third light pattern in which the first light pattern and the second light pattern reflected from an object overlap, so as to separate the third light pattern into a first filtered pattern and a second filtered pattern; a step of localizing the first filtered pattern and classifying first feature points of the first filtered pattern; a step of localizing the second filtered pattern and classifying second feature points of the second filtered pattern; and a step of obtaining position information of the object on the basis of the first feature points and the second feature points.

Abstract: An electronic apparatus is provided. The electronic apparatus includes a camera, a memory in which a plurality of captured images obtained through the camera and a parameter value of the camera are stored, and a processor electrically connected to the camera and the memory. The processor is configured to identify a scene type corresponding to each captured image of the plurality of captured images; extract a feature point of each captured image of the plurality of captured images based on a feature point extraction method corresponding to the identified scene type of each captured image; obtain a calibration parameter value corresponding to a feature type of each extracted feature point; obtain an integrated calibration parameter value based on one or more obtained calibration parameter values; and update the parameter value stored in the memory based on the integrated calibration parameter value.

Abstract: A synchronization method of a first depth sensing apparatus includes: transmitting a first optical signal to measure a distance to an object; receiving the first optical signal reflected by the object; when recognition of the received first optical signal fails, stopping transmission of the first optical signal and generating first synchronization information for synchronization with at least one second depth sensing apparatus; receiving a third optical signal for synchronization with the first depth sensing apparatus, which is transmitted by the at least one second depth sensing apparatus, and decoding the received third optical signal to extract at least one piece of second synchronization information; determining a time point at which and a cycle in which to re-transmit the first optical signal, based on the first synchronization information and the at least one piece of second synchronization information; and re-transmitting the first optical signal at the determined time point and cycle.

Abstract: A three-dimensional depth measuring method using a structured light camera comprising a first projector, a second projector, and an optical sensor, according to the present disclosure, comprises: a step in which a first light pattern is projected by the first projector; a step in which a second light pattern is projected by the second projector; a step of filtering, by a color filter, a third light pattern in which the first light pattern and the second light pattern reflected from an object overlap, so as to separate the third light pattern into a first filtered pattern and a second filtered pattern; a step of localizing the first filtered pattern and classifying first feature points of the first filtered pattern; a step of localizing the second filtered pattern and classifying second feature points of the second filtered pattern; and a step of obtaining position information of the object on the basis of the first feature points and the second feature points.

Abstract: A method and device are provided for obtaining a 3D image. The method includes sequentially projecting a plurality of beams to an object, each of the plurality of projected beams corresponding to a respective one of a plurality of sectors included in a pattern; detecting a plurality of beams reflected off of the object corresponding to the plurality of projected beams; identifying time-of-flight (ToF) of each of the plurality of projected beams based on the plurality of detected beams; identifying a distortion of the pattern, which is caused by the object, based on the plurality of detected beams; and generating a depth map for the object based on the distortion of the pattern and the ToF of each of the plurality of projected beams, wherein the plurality of detected beams are commonly used to identify the ToF and the distortion of the pattern.

Abstract: A method and device are provided for obtaining a 3D image. The method includes sequentially projecting a plurality of beams to an object, each of the plurality of projected beams corresponding to a respective one of a plurality of sectors included in a pattern; detecting a plurality of beams reflected off of the object corresponding to the plurality of projected beams; identifying time-of-flight (ToF) of each of the plurality of projected beams based on the plurality of detected beams; identifying a distortion of the pattern, which is caused by the object, based on the plurality of detected beams; and generating a depth map for the object based on the distortion of the pattern and the ToF of each of the plurality of projected beams, wherein the plurality of detected beams are commonly used to identify the ToF and the distortion of the pattern.