There is provided an image processing apparatus comprising. An obtainment unit obtains a first circular fisheye image accompanied by a first missing region in which no pixel value is present. A generation unit generates a first equidistant cylindrical projection image by performing first equidistant cylindrical transformation processing based on the first circular fisheye image. The generation unit generates the first equidistant cylindrical projection image such that a first corresponding region corresponding to the first missing region has a pixel value in the first equidistant cylindrical projection image.

An image processor and a method therein to provide a target image for evaluation with an object detector. The method comprises: obtaining a source image captured by a camera and depicting an object, and applying an inverse pixel transform to each target pixel of a target image to determine one or more source pixels located at a position in the source image corresponding to a position of each target pixel in the target image. The method further comprises assigning, to each target pixel, a target pixel value determined based on one or more source pixel values of the determined one or more source pixels located at the corresponding position, thereby is a size c in the target image of the depicted object of a specific object type normalized in at least one size dimension. Thereafter, the target image is fed to an object detector for evaluation.

An image device utilizing non-planar projection images to generate a depth map includes two image capturers and a depth engine. The two image capturers are used for generating two non-planar projection images, wherein when each non-planar projection image of the two non-planar projection images is projected to a space corresponding to an image capturer corresponding to the each non-planar projection image, projection positions of each row of pixels of the each non-planar projection image in the space and optical centers of the two image capturers share a plane. The depth engine is used for generating a depth map according to the two non-planar projection images.

The present invention relates to the field of panoramic cameras, and provides a method for tracking a target in a panoramic video, and a panoramic camera. The method is used to track a target in a panoramic video on the basis of a multi-scale correlation filter, and employs automatic electronic pan-tilt-zoom technology. The present invention provides more robust tracking, faster processing speeds, a greater range of application in different tracking scenarios, and can be used to ensure that a tracked target is always at the center of the screen.

Examples are disclosed relating to applying an analytical geometric projection that has been modified by an amplitude function. One example provides a computing device comprising a logic subsystem and a storage subsystem holding instructions executable by the logic subsystem to receive an image of a scene as acquired by an image sensor, apply a mapping to the image of the scene that maps pixels of the image to projected pixels on an analytical projection that is modified by an amplitude function such that the analytical projection achieves a higher zoom effect on pixels closer to a center of the image compared to pixels closer to an edge of the image, thereby obtaining a corrected image, and output the corrected image.

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 360 image data processing method performed by a 360 video receiving device, according to the present invention, comprises the steps of: receiving 360 image data; acquiring information and metadata on an encoded picture from the 360 image data; decoding the picture on the basis of the information on the encoded picture; and rendering the decoded picture and an overlay on the basis of the metadata, wherein the metadata includes overlay-related metadata, the overlay is rendered on the basis of the overlay-related metadata, and the overlay-related metadata includes information on a region of the overlay.

Provided is a display control apparatus including a vehicle detector and a processor, in which the processor performs projection conversion for projecting the plurality of captured images onto a projection plane, performs a visual point conversion process for converting at least one of the plurality of captured images into an image viewed from a virtual visual point, and sets a ground height of the virtual visual point when the surrounding vehicle is not detected in a predetermined region to be lower than a ground height of the virtual visual point when the surrounding vehicle is detected in the predetermined region.

The present invention belongs to the technical field of 3D reconstruction in the field of computer vision, and provides a dataset generation method for self-supervised learning scene point cloud completion based on panoramas. Pairs of incomplete point cloud and target point cloud with RGB information and normal information can be generated by taking RGB panoramas, depth panoramas and normal panoramas in the same view as input for constructing a self-supervised learning dataset for training of the scene point cloud completion network. The key points of the present invention are occlusion prediction and equirectangular projection based on view conversion, and processing of the stripe problem and point-to-point occlusion problem during conversion. The method of the present invention includes simplification of the collection mode of the point cloud data in a real scene; occlusion prediction idea of view conversion; and design of view selection strategy.

Systems and methods for projecting a multi-faceted image onto a convex polyhedron based on an input image are described. A system can include a controller configured to determine a mapping between pixels within a wide-angle image and a multi-faceted image, and generate the multi-faceted image based on the mapping.