A video may include a capture of a scene, such as a wide-field of view capture of the scene. A punchout of the video may provide a framing of the captured scene. The punchout may be determined based on the context of the video, such as the type of captured scene within the video, the motion of the image capture device that captured the video, and/or the motion of one or more things within the captured scene.

An information processing apparatus includes circuitry. The circuitry is configured to detect at least one line segment in a reference direction from a partial area of an image. The circuitry is configured to calculate an inclination of a plane including coordinates of two points included in the at least one line segment in a spherical coordinate system and a reference point of the spherical coordinate system. The circuitry is configured to correct a zenith of the image based on the inclination of the plane.

This invention provides a system and method that efficiently detects objects imaged using a 3D camera arrangement by referencing a cylindrical or spherical surface represented by a point cloud, and measures variant features of an extracted object including volume, height, and center of mass, bounding box, and other relevant metrics. The system and method, advantageously, operates directly on unorganized and un-ordered points, requiring neither a mesh/surface reconstruction nor voxel grid representation of object surfaces in a point cloud. Based upon a cylinder/sphere reference model, an acquired 3D point cloud is flattened. Object (blob) detection is carried out in the flattened 3D space, and objects are converted back to the 3D space to compute the features, which can include regions that differ from the regular shape of the cylinder/sphere. Downstream utilization devices and/or processes, such as part reject mechanism and/or robot manipulators can act on the identified feature data.

Apparatus and methods for providing a frame packing arrangement for the encoding/decoding of, for example, panoramic content. In one embodiment, the frame packing arrangement utilizes overlapping imaging data so as to enable, for example, a post-decode stitching operation to be performed. The frame packing arrangement may utilize a number of projection formats, such as a cubemap projection, and may utilize any number of differing aspect ratios such as, without limitation, 4×3, 3×2, 4×2, 2×4 aspect ratios. Additionally, the overlapping imaging data may be positioned within the frame packing arrangement chosen so as to improve upon coding efficiency. For example, the overlapping imaging data may be positioned within the frame packing arrangement so as to emphasize image continuity. An encoder apparatus and decoder apparatus for use with the aforementioned frame packing arrangements are also disclosed.

The invention relates to an apparatus for generating or processing an image signal. A first image property pixel structure is a two-dimensional non-rectangular pixel structure representing a surface of a view sphere for the viewpoint. A second image property pixel structure is a two-dimensional rectangular pixel structure and is generated by a processor (305) to have a central region derived from a central region of the first image property pixel structure and at least a first corner region derived from a first border region of the first image property pixel structure. The first border region is a region proximal to one of an upper border and a lower border of the first image property pixel structure. The image signal is generated to include the second image property pixel structure and the image signal may be processed by a receiver to recover the first image property pixel structure.

An image capture apparatus capable of eliminating wasteful photographing in predetermined panoramic photography. Predetermined panoramic photography combines images acquired by repeating still image capturing with long time-exposure and panning. A first angle of view in panning direction indicates an angle from a start to an end point in the photography. A photographing condition is calculated based on the first angle and a focal length of a photographic lens, and displayed changeable by a user. When combining the images, geometric deformation processing is performed thereon for changing a second angle of view indicative of an angle of view in a direction orthogonal to the panning direction according to the focal length. The second angle changed by the processing is calculated according to the set first angle and the displayed photographing condition. The second angle is displayed before and after the change.

In a method and apparatus for processing video data, one or more processors are configured to encode a portion of stored video data in a pixel domain to generate pixel domain video data, a first graphics processing unit is configured to process the video data in a graphics domain to generate graphics domain video data, and an interface transmits the graphics domain video data and the pixel domain video data. One or more processors are configured to parse the video data into a graphics stream and an audio-video stream and decode the video data, a sensor senses movement adaptations of a user, and a second graphics processing unit is configured to generate a canvas on a spherical surface with texture information received from the graphics stream, and render a field of view based on the sensed movement adaptations of the user.

A method, according to one example, includes receiving a 360-degree image that was captured by a 360-degree camera, converting the 360-degree image into a rectangular image, and copying an edge portion from a first edge of the rectangular image and pasting the copied edge portion to a second edge of the rectangular image, thereby forming a modified rectangular image. The method further includes applying a neural network to the modified rectangular image to identify objects appearing in the modified rectangular image, wherein the modified rectangular image facilitates object identification near the second edge.

A method, according to one example, includes receiving a 360-degree image that was captured by a 360-degree camera, converting the 360-degree image into a rectangular image, and copying an edge portion from a first edge of the rectangular image and pasting the copied edge portion to a second edge of the rectangular image, thereby forming a modified rectangular image. The method further includes applying a neural network to the modified rectangular image to identify objects appearing in the modified rectangular image, wherein the modified rectangular image facilitates object identification near the second edge.

A video may include a capture of a scene, such as a wide-field of view capture of the scene. A punchout of the video may provide a framing of the captured scene. The punchout may be determined based on the context of the video, such as the type of captured scene within the video, the motion of the image capture device that captured the video, and/or the motion of one or more things within the captured scene.