Systems and methods for image processing for determining a registration map between a first image of a scene with a second image of the scene, include solving an optimal transport (OT) problem to produce the registration map by optimizing a cost function that determines a minimum of a ground cost distance between the first and the second images modified with an epipolar geometry-based regularizer including a distance that quantifies the violation of an epipolar geometry constraint between corresponding points defined by the registration map. The ground cost compares a ground cost distance of features extracted within the first image with a ground cost distance of features extracted from the second image.

A method, system, and computer product that track scanning data acquired by a three-dimensional (3D) coordinate scanner is provided. The method includes storing a digital representation of an environment in memory of a mobile computing device. A first scan is performed with the 3D coordinate scanner in an area of the environment. A location of the first scan is determined on the digital representation. The first scan is registered with the digital representation. The location of the 3D coordinate scanner is indicated on the digital representation at the time of the first scan.

In one implementation, a method includes: obtaining a first depth estimation characterizing a distance between the device and a surface in a real-world environment, wherein the first depth estimation is derived from image data including a representation of the surface; receiving, using the audio transceiver, an acoustic reflection of an acoustic wave, wherein the acoustic wave is transmitted in a known direction relative to the device; and determining a second depth estimation based on the acoustic reflection, wherein the second depth estimation characterizes the distance between the device and the surface in the real-world environment; and determining a confirmed depth estimation characterizing the distance between the device and the surface based on resolving any mismatch between the first depth estimation and the second depth estimation.

An improved UAV system and methods for operation in an inventory management system. The methods include generating a three dimensional (3D) map and estimating a position and orientation of the UAV based upon this map; autonomously navigating the UAV in the environment by using the generated 3d map in conjunction with the position and the orientation of the UAV; performing static and dynamic obstacle avoidance in the environment using collision avoidance; and finding the optimal path from a source node to a destination node within the environment.

A first frame of time of flight (TOF) data including projection off data and infrared data is obtained, and after determining that a data block satisfying that a number of data points with values greater than a first threshold is greater than a second threshold is present in the infrared data, TOF data for generating a first frame of a TOF image is obtained based on a difference between the infrared data and the projection off data. Because the data block satisfying the number of data points with values greater than the first threshold is greater than the second threshold is an overexposed data block, and the projection off data is TOF data acquired by a TOF camera with a TOF light source being off, the difference between the infrared data and the projection off data can correct the overexposure, improving quality of the first frame of the TOF image.

Disclosed is a depth sensor for determining depth. The depth sensor can include a photosensor, a metalens configured to manipulate light to simultaneously produce at least two images having different focal distances on a surface of the photosensor, and processing circuitry configured to receive, from the photosensor, a measurement of the at least two images having different focal distances. The depth sensor can determine, according to the measurement, a depth associated with at least one feature in the at least two images.

An intraoral scanner includes an image capturing device and a processor. A method of operating the intraoral scanner includes the image capturing device sequentially capturing M images of a buccal bite, the processor generating M sets of buccal bite point clouds according to the M images, the processor matching the M sets of buccal bite point clouds to generate a bite model, when the number of data points of the bite model exceeds a first threshold, the processor computing P sets of bite feature descriptors of the bite model, when a predetermined quantity of bite feature descriptors in a set of bite feature descriptors of the P sets of bite feature descriptors exceeds a second threshold, the processor performing a registration on an upper arch model and a lower arch model to the buccal bite mode to generate a full mouth model.

Systems and methods for semantic labeling of point clouds using images. Some implementations may include obtaining a point cloud that is based on lidar data reflecting one or more objects in a space; obtaining an image that includes a view of at least one of the one or more objects in the space; determining a projection of points from the point cloud onto the image; generating, using the projection, an augmented image that includes one or more channels of data from the point cloud and one or more channels of data from the image; inputting the augmented image to a two dimensional convolutional neural network to obtain a semantic labeled image wherein elements of the semantic labeled image include respective predictions; and mapping, by reversing the projection, predictions of the semantic labeled image to respective points of the point cloud to obtain a semantic labeled point cloud.

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.

A point cloud capture system is provided to detect and correct data density during point cloud generation. The system obtains data points that are distributed within a space and that collectively represent one or more surfaces of an object, scene, or environment. The system computes the different densities with which the data points are distributed in different regions of the space, and presents an interface with a first representation for a first region of the space in which a first subset of the data points are distributed with a first density, and a second representation for a second region of the space in which a second subset of the data points are distributed with a second density.