A method, a system, a device, and a computer readable storage medium for video reconstruction are disclosed. The method includes: obtaining an image combination of multi-angle free-perspective video frames, parameter data corresponding to the image combinations of the video frames, and position information of a virtual viewpoint based on a user interaction; selecting texture images and depth maps of corresponding groups in the image combinations of the video frames at a time moment of the user interaction according to a preset rule and based on the position information of the virtual viewpoint and the parameter data corresponding to the image combinations of the video frames; and combining and rendering the texture images and the depth maps of the corresponding groups based on the position information of the virtual viewpoint and parameter data corresponding to the depth maps and the texture images of the corresponding groups to obtain a reconstructed image.

A photographing method implemented by an electronic device includes receiving a photographing instruction of a user, shooting a primary camera image using a primary camera, shooting a wide-angle image using a wide-angle camera, and generating a three-dimensional (3D) image based on the primary camera image and the wide-angle image, where the 3D image includes a plurality of frames of images that is converted from a 3D viewing angle and corresponds to different viewing angles.

A photographing method implemented by an electronic device includes receiving a photographing instruction of a user, shooting a primary camera image using a primary camera, shooting a wide-angle image using a wide-angle camera, and generating a three-dimensional (3D) image based on the primary camera image and the wide-angle image, where the 3D image includes a plurality of frames of images that is converted from a 3D viewing angle and corresponds to different viewing angles.

A method for training a depth estimation model with depth hints is disclosed. For each image pair: for a first image, a depth prediction is determined by the depth estimation model and a depth hint is obtained; the second image is projected onto the first image once to generate a synthetic frame based on the depth prediction and again to generate a hinted synthetic frame based on the depth hint; a primary loss is calculated with the synthetic frame; a hinted loss is calculated with the hinted synthetic frame; and an overall loss is calculated for the image pair based on a per-pixel determination of whether the primary loss or the hinted loss is smaller, wherein if the hinted loss is smaller than the primary loss, then the overall loss includes the primary loss and a supervised depth loss between depth prediction and depth hint. The depth estimation model is trained by minimizing the overall losses for the image pairs.

A method for training a depth estimation model with depth hints is disclosed. For each image pair: for a first image, a depth prediction is determined by the depth estimation model and a depth hint is obtained; the second image is projected onto the first image once to generate a synthetic frame based on the depth prediction and again to generate a hinted synthetic frame based on the depth hint; a primary loss is calculated with the synthetic frame; a hinted loss is calculated with the hinted synthetic frame; and an overall loss is calculated for the image pair based on a per-pixel determination of whether the primary loss or the hinted loss is smaller, wherein if the hinted loss is smaller than the primary loss, then the overall loss includes the primary loss and a supervised depth loss between depth prediction and depth hint. The depth estimation model is trained by minimizing the overall losses for the image pairs.

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.

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.

Examples are disclosed that relate to motion blur corrections for time-of-flight (ToF) depth imaging. One example provides a depth camera comprising a ToF image sensor, a logic machine, and a storage machine storing instructions executable by the logic machine to receive depth image data from the ToF image sensor, the depth image data comprising phase data and active brightness (AB) data, determine a first two-dimensional (2D) AB image corresponding to a first modulation frequency, and determine a second 2D AB image corresponding to a second modulation frequency. The instructions are further executable to determine a 2D translation based upon a comparison between the first 2D AB image and the second 2D AB image, determine corrected phase data based on the 2D translation to form corrected phase data, perform phase unwrapping on the corrected phase data to obtain a three-dimensional (3D) depth image, and output the 3D depth image.

Examples are disclosed that relate to motion blur corrections for time-of-flight (ToF) depth imaging. One example provides a depth camera comprising a ToF image sensor, a logic machine, and a storage machine storing instructions executable by the logic machine to receive depth image data from the ToF image sensor, the depth image data comprising phase data and active brightness (AB) data, determine a first two-dimensional (2D) AB image corresponding to a first modulation frequency, and determine a second 2D AB image corresponding to a second modulation frequency. The instructions are further executable to determine a 2D translation based upon a comparison between the first 2D AB image and the second 2D AB image, determine corrected phase data based on the 2D translation to form corrected phase data, perform phase unwrapping on the corrected phase data to obtain a three-dimensional (3D) depth image, and output the 3D depth image.

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.