A computer system generates stereo image data from monocular images. The system generates depth maps for single images using a monocular depth estimation method. The system converts the depth maps to disparity maps and uses the disparity maps to generate additional images forming stereo pairs with the monocular images. The stereo pairs can be used to form a stereo image training data set for training various models, including depth estimation models or stereo matching models.

Techniques for rendering a 3D virtual object in an augmented-reality system are described. A system, a method, and a non-transitory memory device for augmented reality rendering of three-dimensional, virtual objects are described. In an example, a number of images of an environment are acquired; relative movement of a camera acquiring the number of images is tracked; camera pose is determined relative to the environment using the number of images and tracked relative movement of the camera; depth and normal surfaces of objects in the environment are estimated using a depth map and a normal map; a surface geometry of the environment is reconstructed using the depth map and the normal map; and the virtual object is rendered using the surface geometry of the environment.

Techniques for rendering a 3D virtual object in an augmented-reality system are described. A system, a method, and a non-transitory memory device for augmented reality rendering of three-dimensional, virtual objects are described. In an example, a number of images of an environment are acquired; relative movement of a camera acquiring the number of images is tracked; camera pose is determined relative to the environment using the number of images and tracked relative movement of the camera; depth and normal surfaces of objects in the environment are estimated using a depth map and a normal map; a surface geometry of the environment is reconstructed using the depth map and the normal map; and the virtual object is rendered using the surface geometry of the environment.

Methods, apparatuses, mediums, and devices for generating multi-angle free-respective image data are provided. The method for generating multi-angle free-perspective image data includes: acquiring multiple synchronized images, where the multiple images have different shooting angles; determining the depth data of each image based on the multiple images; and for each of the images, storing pixel data of the image in a first field and storing depth data in a second field associated with the first field. Technical solutions in the example embodiments of the present invention may improve the user experience.

Methods, apparatuses, mediums, and devices for generating multi-angle free-respective image data are provided. The method for generating multi-angle free-perspective image data includes: acquiring multiple synchronized images, where the multiple images have different shooting angles; determining the depth data of each image based on the multiple images; and for each of the images, storing pixel data of the image in a first field and storing depth data in a second field associated with the first field. Technical solutions in the example embodiments of the present invention may improve the user experience.

Methods and systems are provided for a reprojection engine for augmented-reality devices. The augmented-reality device projects virtual content within a real-world environment. The augmented-reality device tracks a six degrees of freedom headpose of the augmented-reality device, depth information of the virtual content, motion vectors that correspond to movement of the virtual content, and a color buffer for a reprojection engine. The reprojection engine generates a reprojection of the virtual content defined by an extrapolation of a first frame using the headpose, the depth information, motion vectors, and the color surface data structure. The reprojected virtual content continues to appear as if positioned with the real-world environment regardless of changes in the headpose of the augmented-reality device or motion of the virtual content.

A stereoscopic visualization system using shape from shading algorithm is an image conversion device connected between a monoscopic endoscope and a 3D monitor. The system applies the algorithm which generates a depth map for a 2D image of video frames. The algorithm first calculates a direction of a light source for the 2D image. Based upon the information of light distribution and shading for the 2D image, the depth map is generated. The depth map is used to calculate another view of the original 2D image by depth image based rendering algorithm in generation of stereoscopic images. After the new view is rendered, the stereoscopic visualization system also needs to convert the display format of the stereoscopic images for different kinds of 3D displays. Based on this method, it can replace the whole monoscopic endoscope with a stereo-endoscope system and no modification is required for the monoscopic endoscope.

A stereoscopic visualization system using shape from shading algorithm is an image conversion device connected between a monoscopic endoscope and a 3D monitor. The system applies the algorithm which generates a depth map for a 2D image of video frames. The algorithm first calculates a direction of a light source for the 2D image. Based upon the information of light distribution and shading for the 2D image, the depth map is generated. The depth map is used to calculate another view of the original 2D image by depth image based rendering algorithm in generation of stereoscopic images. After the new view is rendered, the stereoscopic visualization system also needs to convert the display format of the stereoscopic images for different kinds of 3D displays. Based on this method, it can replace the whole monoscopic endoscope with a stereo-endoscope system and no modification is required for the monoscopic endoscope.

A method and a device for processing multi-view video data. The multi-view video data includes at least one part of a decoded image of at least one view of the multi-view video, from an encoded data stream representative of the multi-view video. At least one item of data, referred to as synthesis data, is obtained from at least the one part of the decoded image, and at least one image of an intermediate view of the multi-view video not encoded in the encoded data stream is synthesized from at least the one part of the decoded image and from the synthesis data obtained.

Embodiments disclose an augmented reality-based remote guidance method and apparatus, terminal, and storage medium. The method comprises the following steps: acquiring a two-dimensional video of a target scene, and sending the two-dimensional video to a remote terminal; if a guidance mode of the remote guidance is marking mode, acquiring two-dimensional pixel coordinates corresponding to a marked point in a marked image frame of the two-dimensional video at the remote terminal; determining current camera coordinates corresponding to the marked point, according to first three-dimensional coordinate estimation rules and the two-dimensional pixel coordinates, wherein the current camera coordinates are current three-dimensional space coordinates corresponding to the marked point in a camera coordinate system; and according to a presentation mode and the current camera coordinates rendering a three-dimensional virtual model corresponding to the marked point so as to display the three-dimensional virtual model in the target scene.