Systems and methods for generating a three-dimensional reconstruction from a two-dimensional image of an asset are described. In some aspects, at least one computer hardware processor is used to perform: receiving a two-dimensional input image of an asset; encoding, using a first statistical model, the input image into a latent vector; and generating, using a second statistical model, at least one depth image from the latent vector, wherein pixel values in the at least one depth image correspond to coordinates of a point cloud.

Systems and methods for generating a three-dimensional reconstruction from a two-dimensional image of an asset are described. In some aspects, at least one computer hardware processor is used to perform: receiving a two-dimensional input image of an asset; encoding, using a first statistical model, the input image into a latent vector; and generating, using a second statistical model, at least one depth image from the latent vector, wherein pixel values in the at least one depth image correspond to coordinates of a point cloud.

A system and method are disclosed for, using structure-from-motion techniques, projecting a building information model (BIM) into images from photographs taken of a construction site, to generate a 3D point cloud model using the BIM and, when combined with scheduling constraints, facilitates 4D visualizations and progress monitoring. One of the images acts as an anchor image. Indications are received of first points in the anchor image that correspond to second points in the BIM. Calibration information for an anchor camera is calculated based on the indications and on metadata extracted from the anchor image, to register the anchor image in relation to the BIM. A homography transformation is determined between the images and the anchor camera using the calibration information, to register the rest of the images with the BIM, where some of those images are taken from different cameras and from different angles to the construction site.

An imaging quality control system (80) employing an imaging quality controller (84) and a monitor (81). In operation, the imaging quality controller (84) executes an image processing of subject image data of the anatomical object (e.g., subject non-segmentation-based or segmentation-based image registration of US, CT and/or MRI anatomical images), and assessing an accuracy of the image processing of the subject image data of the anatomical object as a function of a subject Eigen weight set relative to a training Eigen range set (e.g., previously registered or segmented US, CT and/or MRI anatomical images). The subject Eigen weight set is derived from the subject image data of the anatomical object, and the training Eigen range set is derived from training image data of anatomical object. The monitor (81) displays the assessment of the accuracy of the image processing of the subject image data of the anatomical object by the imaging quality controller (84).

A method generates and uses a correspondence probability map for visualization of two image datasets. The method includes obtaining two image datasets and obtaining an image registration algorithm that includes a correspondence model. The method further includes registering the two image datasets to generate a displacement vector field and generating a correspondence probability map, using the correspondence model, based on the two image datasets. The method further includes using the correspondence probability map to visualize the two image datasets. A computing system (120) includes a memory device (124) configured to store instructions, including a visualization module (130), and a processor (122) that executes the instructions, which causes the processor to generate and employ a correspondence probability map for visualization of two image datasets.

A computer-implemented method for generating a bird’s eye view image of a scene includes: (a) acquiring at least one lidar frame comprising points with inherent distance information and at least one camera image of the scene; (b) generating a mesh representation of the scene by using the at least one lidar frame, the mesh representation representing surfaces shown in the scene with inherent distance information; (c) generating a mask image by classifying pixels of the at least one camera image as representing ground pixels or non-ground pixels of the at least one camera image; and (d) generating the bird’s eye view image by enhanced inverse perspective mapping exploiting distance information inherent to the surfaces of the mesh representation, pixels of the mask image classified as ground pixels, and the at least one camera image.

A computer-implemented method for generating a bird’s eye view image of a scene includes: (a) acquiring at least one lidar frame comprising points with inherent distance information and at least one camera image of the scene; (b) generating a mesh representation of the scene by using the at least one lidar frame, the mesh representation representing surfaces shown in the scene with inherent distance information; (c) generating a mask image by classifying pixels of the at least one camera image as representing ground pixels or non-ground pixels of the at least one camera image; and (d) generating the bird’s eye view image by enhanced inverse perspective mapping exploiting distance information inherent to the surfaces of the mesh representation, pixels of the mask image classified as ground pixels, and the at least one camera image.

Techniques are described for determining a pre-morbid shape of an anatomical object. A method includes receiving first image data of a first anatomical structure and second image data of a second anatomical structure. The first and second anatomical structures are anatomically related. The method includes determining a first shape model based on the first image data and a joint statistical shape model (SSM). The method includes determining a second shape model based on the first shape model, the first image data, and the second image data, the second shape model including a second estimated shape of the first anatomical structure and a second estimated shape for the second anatomical structure. The method includes generating anatomical information indicative of the pre-morbid shape of at least the second anatomical structure based on the second shape model.

Techniques are described for determining a pre-morbid shape of an anatomical object. A method includes receiving first image data of a first anatomical structure and second image data of a second anatomical structure. The first and second anatomical structures are anatomically related. The method includes determining a first shape model based on the first image data and a joint statistical shape model (SSM). The method includes determining a second shape model based on the first shape model, the first image data, and the second image data, the second shape model including a second estimated shape of the first anatomical structure and a second estimated shape for the second anatomical structure. The method includes generating anatomical information indicative of the pre-morbid shape of at least the second anatomical structure based on the second shape model.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for image splicing are provided. In one aspect, a method includes obtaining at least two groups of to-be-spliced images and scanning information corresponding to the at least two groups of to-be-spliced images, determining an overlap area of the at least two groups of to-be-spliced images and a physical sequence of splicing according to the scanning information, registering the at least two groups of to-be-spliced images based on the overlap area to obtain a registration result, and splicing the at least two groups of registered to-be-spliced images according to the physical sequence of splicing and the registration result.