An example wear detection system receives first image data related to at least one ground engaging tool (GET) of a work machine from one or more sensors at a first time instance in a dig-dump cycle of the work machine. The wear detection system processes the first image data to determine a first wear measurement and first wear level for the at least one GET. The wear detection system determines whether the first wear level is indicative of a GET replacement condition. The wear detection system generates an alert when the first wear level is indicative of the GET replacement condition. The wear detection system receives second image data related to the at least one GET a second time instance different from the first time instance when the first wear level is not indicative of the GET replacement condition and determines a second wear measurement and second wear level for the at least one GET. The wear detection system generates an alert indicative of the first wear level and the second wear level based on determining that the first wear level and the second wear level are indicative of the GET replacement condition.

A solution to the problem of image and video stitching is disclosed that compensates for the effects of lens distortion, camera misalignment, and parallax in combining multiple images. The disclosed image stitching technique includes depth or disparity estimation, alignment, and blending processes configured to be computationally efficient and produce quality results by limiting the presence of noticeable seams and artifacts in the final stitched image. An inter-frame approach applies image stitching to video frames to maintain temporal continuity between successive frames across a stitched video output having a 360-degree viewing perspective. A temporal adjustment is configured to improve temporal continuity between a subsequent frame and a previous frame in a sequence of video frames.

A system for dense depth computation aided by sparse feature matching generates a first image using a first camera, a second image using a second camera, and a third image using a third camera. The system generates a sparse disparity map using the first image and the third image by (1) identifying a set of feature points within the first image and a set of corresponding feature points within the third image, and (2) identifying feature disparity values based on the set of feature points and the set of corresponding feature points. The system also applies the first image, the second image, and the sparse disparity map as inputs for generating a dense disparity map.

An image-based method of modeling and rendering a three-dimensional model of an object is provided. The method comprises: obtaining a three-dimensional point cloud at each frame of a synchronized, multi-view video of an object, wherein the video comprises a plurality of frames; extracting a feature descriptor for each point in the point cloud for the plurality of frames without storing the feature descriptor for each frame; producing a two-dimensional feature map for a target camera; and using an anti-aliased convolutional neural network to decode the feature map into an image and a foreground mask.

Aspects of the present disclosure involve a system for presenting AR items. The system performs operations including: receiving an image that includes a depiction of a first real-world body part in a real-world environment; applying a machine learning technique to the image to generate a plurality of dense outputs each associated with a respective pixel of a plurality of pixels in the image; applying a first task-specific decoder to the plurality of dense outputs to identify a pixel corresponding to a center of the first real-world body part; applying a second task-specific decoder using the identified pixel to retrieve a 3D rotation, translation and scale of first real-world body part from the plurality of dense outputs; modifying an AR object based on the 3D rotation, translation, and scale of first real-world body part; and modifying the image to include a depiction of the modified AR object.

Disclosed are systems, methods, and non-transitory computer-readable media for misaligned vantage point mitigation for computer stereo vision. A misaligned vantage point mitigation system determines whether vantage points of the optical sensors are misaligned from an expected vantage point and, if so, determines an adjustment variable to mitigate the misalignment based on the location of matching features identified in images captured by both optical sensors.

Input images are partitioned into non-overlapping segments perpendicular to a disparity dimension of the input images. Each segment includes a contiguous region of pixels spanning from a first edge to a second edge of the image, with the two edges parallel to the disparity dimension. In some aspects, contiguous input image segments are assigned in a “round robin” manner to a set of sub-images. Each pair of input images generates a corresponding pair of sub-image sets. Semi-global matching processes are then performed on pairs of corresponding sub-images generated from each input image. The SGM processes may be run in parallel, reducing an elapsed time to generate respective disparity sub-maps. The disparity sub-maps are then combined to provide a single disparity map of equivalent size to the original two input images.

Described herein are system and methods to improve the image quality of a multiview image. In some embodiments a zero disparity plane image is generated based on a view of a multiview by identifying portions of the multiview image that correspond to the zero disparity plane. The zero disparity plane image and view images of the multiview image may be transmitted to a time-multiplexed display. The time-multiplexed display may operate according to a two-dimensional (2D) mode and a multiview mode. The time-multiplexed display may be configured to display the zero disparity plane image and the view images as a composite image.

Embodiments of the present invention disclose a binocular image matching method, apparatus, device, and storage medium. The method comprises: performing target detection on a first image to obtain a first bounding box of a target in the first image; determining a second bounding box corresponding to the first bounding box in a second image; and obtaining a third bounding box of the target in the second image by regressing the second bounding box. The technical solutions realize accurate matching of targets in a binocular image without requiring performing target detection on both images in the binocular image, and then use a matching algorithm to match the targets detected in the two images, thereby greatly reducing the calculation overhead of target matching in the binocular image.

Systems and techniques are described for processing image data to generate an image with a synthetic depth of field (DoF). An imaging system receives first image data of a scene captured by a first image sensor. The imaging system receives second image data of the scene captured by a second image sensor. The first image sensor is offset from the second image sensor by an offset distance. The imaging system generates, using at least the first image data and the second image data as inputs to one or more trained machine learning systems, an image having a synthetic depth of field corresponding to a simulated aperture size. The simulated aperture size is associated with the offset distance. The imaging system outputs the image.