Provided is a computer-implemented method of training a neural network model by augmenting images representing objects. The method includes: obtaining a first object recognition result predicted by a first neural network model using, as an input, a first image captured by a first camera capturing, from a first viewpoint, a space including at least one object; converting the obtained first object recognition result, based on a conversion relationship between a first camera coordinate system corresponding to the first camera and a second camera coordinate system corresponding to a second camera capturing, from a second viewpoint, the space; generating, based on the first object recognition result converted with respect to the second viewpoint, training data by performing labeling on a second image that corresponds to the first image, the second image being captured by the second camera; and training a second neural network model by using the generated training data.

The invention relates to a pinhole camera comprising a cleaning chamber for maintaining the quality of images taken by the pinhole camera, the pinhole camera comprising a lens having a rectangular cross-section, a cover glass in front of the lens, for preventing the access of washing liquid, air and impurities into the camera. Furthermore, the camera comprises a cleaning chamber formed between the cover glass and the wall comprising the pinhole of the pinhole camera by means of a flange, the flange being provided with an opening from which compressed air and washing liquid can be supplied by a tube into the cleaning chamber, for cleaning the cleaning chamber. The compressed air and the washing liquid are configured to exit the cleaning chamber via the pinhole of the pinhole camera. Furthermore, the invention relates to a web monitoring beam and a system for cleaning the pinhole camera.

The invention relates to a pinhole camera comprising a cleaning chamber for maintaining the quality of images taken by the pinhole camera, the pinhole camera comprising a lens having a rectangular cross-section, a cover glass in front of the lens, for preventing the access of washing liquid, air and impurities into the camera. Furthermore, the camera comprises a cleaning chamber formed between the cover glass and the wall comprising the pinhole of the pinhole camera by means of a flange, the flange being provided with an opening from which compressed air and washing liquid can be supplied by a tube into the cleaning chamber, for cleaning the cleaning chamber. The compressed air and the washing liquid are configured to exit the cleaning chamber via the pinhole of the pinhole camera. Furthermore, the invention relates to a web monitoring beam and a system for cleaning the pinhole camera.

A third model training part trains a second neural network model by backpropagation using an error difference between: an identification result which a third neural network model including a trained first neural network model and the second neural network connected to each other outputs after receiving second sensing data and a first operation parameter, and correct identification information corresponding to the second sensing data. A second operation parameter acquisition part acquires a second operation parameter by updating the first operation parameter via the first neural network model by the backpropagation.

A third model training part trains a second neural network model by backpropagation using an error difference between: an identification result which a third neural network model including a trained first neural network model and the second neural network connected to each other outputs after receiving second sensing data and a first operation parameter, and correct identification information corresponding to the second sensing data. A second operation parameter acquisition part acquires a second operation parameter by updating the first operation parameter via the first neural network model by the backpropagation.

The present disclosure relates to a pinhole camera, an electronic apparatus and a manufacturing method, an electronic apparatus and a manufacturing method by which further reduction in profile can be achieved. The pinhole camera includes a semiconductor substrate on which a pixel array unit on which a plurality of pixels are arranged in an array is formed, a protective substrate stacked on an on-chip lens arranged corresponding to a pixel array unit of the semiconductor substrate through a seal resin, and a light shielding film formed on the protective substrate and configured to block light to be irradiated upon the semiconductor substrate. A pinhole is formed in the light shielding film. The present technology can be applied, for example, to a CMOS image sensor of the stacked type.

To provide a stereo viewer and/or a stereo view finder that enables a stereo image displayed on an electronic display to be easily visualized and to enhance portability so as to be easily and conveniently carried when photographing and viewing the stereo image outside. There is provided a stereo viewer/stereo view finder (1) including a foldable light shielding hood (10) attached on a liquid crystal display (D), and a magnifying lens (20) including a pair of lenses (21R, 21L) for viewing stereo images (SR, SL) and a lens folder (22), the magnifying lens being attached to a front inner wall surface (11a) of the light shielding hood (10) by way of a hinge (23) and being developable to be parallel to the liquid crystal display (D).

To provide a stereo viewer and/or a stereo view finder that enables a stereo image displayed on an electronic display to be easily visualized and to enhance portability so as to be easily and conveniently carried when photographing and viewing the stereo image outside. There is provided a stereo viewer/stereo view finder (1) including a foldable light shielding hood (10) attached on a liquid crystal display (D), and a magnifying lens (20) including a pair of lenses (21R, 21L) for viewing stereo images (SR, SL) and a lens folder (22), the magnifying lens being attached to a front inner wall surface (11a) of the light shielding hood (10) by way of a hinge (23) and being developable to be parallel to the liquid crystal display (D).

An optical lens system for photographing an object, according to an embodiment of the present invention, comprises an object image forming member for forming an image of the object, a light source for providing light to the object, optical lenses, an aperture stop area, and a switching aperture disposed in the aperture stop area. The switching aperture includes a base plate having, in the center thereof, an opening for an image to be taken, a blade having, in the center thereof, a pin hole for an image to be magnified such that a magnified image corresponding to the object is formed, and a variable connection member bent or straightened according to power supply to induce light incident from the object to form a photographed image via the opening for an image to be taken or form a magnified image via the pin hole for an image to be magnified.

An optical lens system for photographing an object, according to an embodiment of the present invention, comprises an object image forming member for forming an image of the object, a light source for providing light to the object, optical lenses, an aperture stop area, and a switching aperture disposed in the aperture stop area. The switching aperture includes a base plate having, in the center thereof, an opening for an image to be taken, a blade having, in the center thereof, a pin hole for an image to be magnified such that a magnified image corresponding to the object is formed, and a variable connection member bent or straightened according to power supply to induce light incident from the object to form a photographed image via the opening for an image to be taken or form a magnified image via the pin hole for an image to be magnified.