An apparatus comprising an interface, a structured light projector and a processor. The interface may receive pixel data. The structured light projector may generate a structured light pattern. The processor may process the pixel data arranged as video frames, perform operations using a convolutional neural network to determine a binarization result and an offset value and generate disparity and depth maps in response to the video frames, the structured light pattern, the binarization result, the offset value and a removal of error points. The convolutional neural network may perform a partial block summation to generate a convolution result, compare the convolution result to a speckle value to determine the offset value, generate an adaptive result in response to performing a convolution operation, compare the video frames to the adaptive result to generate the binarization result for the video frames, and remove the error points from the binarization result.

In one embodiment, a method is provided. The method includes obtaining a sequence of images of a three-dimensional volume of a material. The method also includes determining a set of features based on the sequence of images and a first neural network. The set of features indicate microstructure features of the material. The method further includes determining a set of material properties of the three-dimensional volume of the material based on the set of features and a first transformer network.

Embodiments provide systems and methods for three-dimensional building generation from machine learning and topological models. The method uses topology models that are converted into vertices and edges. A BGAN (Building generative adversarial network) is used to create fake vertices/edges. The BGAN is then used to generate random samples from seen sample of different structures of building based on relationship of vertices and edges. The embeddings are then fed into a machine trained network to create a digital structure from the image.

In one embodiment, a computing system may update a first 3D model of a region of an environment based on comparisons between the first 3D model and first depth measurements of the region generated during a first time period. The computing system may determine that the region is static by comparing the first 3D model to second depth measurements of the region generated during a second time period. The computing system may in response to determining that the region is static, detect whether the region changed after the second time period based on comparisons between a second 3D model of the region and third depth measurements of the region generated after the second time period, the second 3D model having a lower resolution than the first 3D model. The computing system may in response to detecting a change in the region, update the first 3D model of the region.

Aspects of the disclosure provide a device for processing frames with aliasing artifacts. For example, the device can include a motion estimation circuit, a warping circuit coupled to the motion estimation circuit, and a temporal decision circuit coupled to the warping circuit. The motion estimation circuit can estimate a motion value between a current frame and a previous frame. The warping circuit can warp the previous frame based on the motion value such that the warped previous frame is aligned with the current frame and determine whether the current frame and the warped previous frame are consistent. The temporal decision circuit can generate an output frame, the output frame including either the current frame and the warped previous frame when the current frame and the warped previous frame are consistent, or the current frame when the current frame and the warped previous frame are not consistent.

A motion detection method includes providing a buffer including a first buffer associated with a background image and a second buffer associated with a foreground image; checking first similarity between the gray level of an input pixel and the first gray level of the first buffer; determining the input pixel as a still pixel if the first similarity is true; checking second similarity between the gray level and the second gray level of the second buffer; determining the input pixel as a moving pixel if the second similarity is false; determining the input pixel as the moving pixel if the second count value is less than the first count value; and determining the input pixel as the still pixel and swapping the first buffer with the second buffer, if the second count value is not less than the first count value.

A method of generating a modified video file using one or more processors is disclosed. The method comprises detecting objects that are represented in an original video file using computer vision object-detection techniques, determining object motion characteristics for the detected objects, based on a specific object motion characteristic for a specific detected object meeting certain requirements, selecting a corresponding audio or visual effect, and applying the corresponding visual effect to the original video file to create the modified video file.

A vehicle generates a city-scale map. The vehicle includes one or more Lidar sensors configured to obtain point clouds at different positions, orientations, and times, one or more processors, and a memory storing instructions that, when executed by the one or more processors, cause the system to perform registering, in pairs, a subset of the point clouds based on respective surface normals of each of the point clouds; determining loop closures based on the registered subset of point clouds; determining a position and an orientation of each of the subset of the point clouds based on constraints associated with the determined loop closures; and generating a map based on the determined position and the orientation of each of the subset of the point clouds.

Methods and systems for detecting visual relations in a video are disclosed. A method comprises: decomposing the video sequence into a plurality of segments; for each segment, detecting objects in frames of the segment; tracking the detected objects over the segment to form a set of object tracklets for the segment; for the detected objects, extracting object features; for pairs of object tracklets of the set of object tracklets, extracting relativity features indicative of a relation between the objects corresponding to the pair of object tracklets; forming relation feature vectors for pairs of object tracklets using the object features of objects corresponding to respective pairs of object tracklets and the relativity features of the respective pairs of object tracklets; and generating a set of segment relation prediction results from the relation features vectors; generating a set of visual relation instances for the video sequence by merging the segment prediction results from different segments; and generating a set of visual relation detection results from the set of visual relation instances.

Methods, apparatus, systems and articles of manufacture are disclosed for augmented reality vehicle condition inspection. An example apparatus disclosed herein includes a location analyzer to determine whether a camera is at an inspection location and directed towards a first vehicle in an inspection profile, the inspection location corresponding to a location of the camera relative to the first vehicle, an interface generator to generate an indication on a display that the camera is at the inspection location, the indication associated with an inspection image being captured, and an image analyzer to compare the inspection image captured at the inspection location with a reference image taken of a reference vehicle of a same type as the first vehicle, and determine a vehicle part condition or a vehicle condition based on the comparison of the inspection image and the reference image.