Abstract: A system permitting one or more users to navigate a wide-angle image, and in particular a panoramic image, using a neural network to correct distortion in that image. To train the neural network, a calibration pattern which defines an array of calibration points is disposed to occupy a field of view of a wide-angle imaging apparatus, and a calibration image is thereby captured by the apparatus. Respective view directions of the calibration points and the positions of these points in the calibration image are used as data for training the neural network to correctly match view directions to corresponding intra-image positions.

Abstract: A panoramic imaging lens having an annular light incident surface formed in a substantial convex lens form; a first reflective surface formed in an annular concave mirror form to reflect light inside the lens; a second reflective surface provided at a central part inside the annular light incident surface to reflect the reflected light from the first reflective surface toward an inner part of the annular first reflective surface; and a light outgoing surface positioned at a central part inside the annular first reflective surface and opposing the second reflective surface to transmit light. A non-reflective part exerting no regular reflection of light is provided on a light path toward the light incident surface amongst light paths of light proceeding to agree with a light path of imaging light incident on and refracted at the light incident surface and proceeding inside the lens.