An apparatus for applying a convolutional neural network (CNN) to a wide-angle camera image is described herein. The apparatus includes a camera, controller, convolution mechanism and a fully connected layer. The camera is to capture a wide-angle image, and the controller is to map the image on a 3D surface. The convolution mechanism is to perform convolution on the 3D surface and the fully connected layer is to classify a plurality of features generated by the convolution mechanism.

A panoramic video stream is compressed. Sinusoidal projection is performed on spherical input images to generate pseudo-cylindrical projection images. A lower-left region and a lower-right region of the image are cut and moved to upper corners of the rectangular bounding box around the pseudo-cylindrical projection image. These upper corners are non-effective areas of default dark pixels with no image pixels. A bottom one-third of the rows of pixels from the image that contained the moved regions are deleted, compressing the image by 33%. Default dark pixels in interface regions between the moved regions and the pseudo-cylindrical projection image are linear interpolated to provide gradual changes in pixel values across the remaining formerly non-effective regions, preventing encoding artifacts caused by abrupt changes in pixel values. Functions may be implemented using lookup tables. Non-focus image areas may be downsampled for additional compression using multi-resolution mapping and adaptive view streaming.

An exemplary video processing method includes: receiving an omnidirectional content corresponding to a sphere; obtaining a plurality of projection faces from the omnidirectional content of the sphere according to a pyramid projection; creating at least one padding region; and generating a projection-based frame by packing the projection faces and the at least one padding region in a pyramid projection layout. The projection faces packed in the pyramid projection layout include a first projection face. The at least one padding region packed in the pyramid projection layout includes a first padding region. The first padding region connects with at least the first projection face, and forms at least a portion of one boundary of the pyramid projection layout.

A video processing method includes receiving a bitstream, and decoding, by a video decoder, the bitstream to generate a decoded frame. The decoded frame is a projection-based frame that has a 360-degree image/video content represented by triangular projection faces packed in a triangle-based projection layout. An omnidirectional image/video content of a viewing sphere is mapped onto the triangular projection faces via a triangle-based projection of the viewing sphere. An equator of the viewing sphere is not mapped along any side of each of the triangular projection faces.

A trailer angle detection system for a vehicle towing a trailer includes a camera disposed at a rear portion of the vehicle and having a field of view rearward of the vehicle. The camera includes a two dimensional imager array having multiple rows and columns of photosensing elements. An image processor processes image data captured by the camera. With the vehicle towing a trailer, and via processing by the image processor of image data captured by the camera, the trailer angle detection system detects a portion of the trailer that is being towed by the vehicle. The trailer angle detection system determines, via processing by the image processor of image data captured by the camera, an angle of the trailer relative to the vehicle responsive to determination of which columns of photosensing elements sense the detected portion of the trailer in the field of view of the camera.

Disclosed is an image generation apparatus generating and outputting a panoramic image that is obtained by converting, to a planar shape, a projection plane onto which a scene within at least a partial range of a virtual sphere as viewed from an observation point is projected. The panoramic image is such that a unit area on the virtual sphere containing a given attention direction as viewed from the observation point is converted to a broader area than other unit areas. The image generation apparatus generates the panoramic image corresponding to the projection plane such that a portion of a main line that links a position in the attention direction to a position in a direction opposite the attention direction and is within the panoramic image corresponding to a unit amount of an angle of rotation around the observation point is maximized in length at a position closest to the attention direction.

An image processing device includes an input unit configured to input a first image; a selector configured to select a second image added to the first image; a converter configured to generate a converted image by converting the second image based on the first image; and a synthesizer configured to generate a synthesized image by synthesizing the first image and the converted image.

A browsing system includes an image distribution apparatus and an information processing apparatus. The image distribution apparatus includes a memory configured to store and register a presentation pattern of a plurality of display positions, including at least a first display position and a second display position, set within a wide-angle image; and first circuitry configured to transmit the wide-angle image and the presentation pattern of the plurality of display positions. The information processing apparatus includes second circuitry configured to receive the wide-angle image and the presentation pattern from the image distribution apparatus, and to display the wide-angle image on a display connected with the information processing apparatus, using the plurality of display positions by applying the received presentation pattern.

An information processing apparatus configured to paste a full-spherical panoramic image along an inner wall of a virtual three-dimensional sphere; calculate an arrangement position for arranging a planar image closer to a center point of the virtual three-dimensional sphere than the inner wall, in such an orientation that a line-of-sight direction from the center point to the inner wall and a perpendicular line of the planar image are parallel to each other, the planar image being obtained by pasting an embedding image to be embedded in the full-spherical panoramic image, on a two-dimensional plane; and display a display image on a display unit. The display image is a two-dimensional image viewed from the center point in the line-of-sight direction in a state in which the full-spherical panoramic image is pasted along the inner wall of the virtual three-dimensional sphere and the planar image is arranged at an arrangement position.

An image processing method and apparatus are provided. The image processing apparatus includes an interface configured to output an input frame and metadata including type information and subtype information; and a rendering unit configured to: determine a type of a polyhedron included in an output frame based on the type information, determine attributes of arrangement of a plurality of areas included in the input frame based on the subtype information, and render the output frame by mapping each of the plurality of areas to corresponding faces of the polyhedron, based on the attributes of arrangement of the plurality of areas.