Described herein are apparatus and methods for orthographic imaging of particles. Particularly, a method to obtain contact-free images of aerosol particles with digital holography from three orthogonal directions is described and demonstrated. Diode lasers of different wavelengths simultaneously illuminate free flowing particles to form holograms on three sensors. Images of the particles are reconstructed from the holograms and used to infer the three-dimensional structure of single spherical particles or clusters of sphere-like particles. The apparatus employs inexpensive components and requires no lenses to achieve the imaging, which gives it a large sensing volume and simple design.

A dimension measurement method includes: extracting a plurality of lines from a plurality of images generated by shooting a target area from a plurality of viewpoints, and generating a line segment model which is a three-dimensional model of the target area that is expressed using the plurality of lines; calculating a dimension of a particular part inside the target area, using the line segment model; and outputting the dimension calculated.

A visual positioning method and apparatus, an electronic device, a computer readable storage medium, and a computer program product are provided. The method includes: performing contour enhancement processing on an actual building image included in a image for positioning to obtain an actual building contour, and determining location information of a target building matching the actual building contour from a preset contour map, the contour map being obtained by blurring non-building contour information in a real panoramic map, and finally generating a visual positioning result based on the location information.

A visual positioning method and apparatus, an electronic device, a computer readable storage medium, and a computer program product are provided. The method includes: performing contour enhancement processing on an actual building image included in a image for positioning to obtain an actual building contour, and determining location information of a target building matching the actual building contour from a preset contour map, the contour map being obtained by blurring non-building contour information in a real panoramic map, and finally generating a visual positioning result based on the location information.

A system includes a sensor and a client. The client receives a set of frames of top-view depth images generated by the sensor. The client identifies a frame of the received frames in which a first contour associated with a first object is merged with a second contour associated with a second object. The client determines, at a first depth in the identified frame, a merged-contour region which is associated with the merged contours. The client detects a third contour at a second depth that is less than the first depth and determines a first region associated with the third contour. The client detects a fourth contour at the second depth and determines a second region associated with the fourth contour. If criteria are satisfied, the client associates the first region with a position of the first object and associates the second region with a position of the second object.

A system includes a sensor and a client. The client receives a set of frames of top-view depth images generated by the sensor. The client identifies a frame of the received frames in which a first contour associated with a first object is merged with a second contour associated with a second object. The client determines, at a first depth in the identified frame, a merged-contour region which is associated with the merged contours. The client detects a third contour at a second depth that is less than the first depth and determines a first region associated with the third contour. The client detects a fourth contour at the second depth and determines a second region associated with the fourth contour. If criteria are satisfied, the client associates the first region with a position of the first object and associates the second region with a position of the second object.

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.

Highly accurate three-dimensional shape data is obtained easily. The information processing apparatus obtains data of a plurality of captured images obtained by capturing an object from a plurality of viewpoints different from one another. Further, the information processing apparatus obtains first three-dimensional shape data indicating the shape of the object, which includes information indicating the spatial position of each of a plurality of feature points indicating features of the object. Furthermore, the information processing apparatus obtains second three-dimensional shape data indicating the shape of the object by using the captured image. Here, the information processing apparatus obtains, at the time of obtaining the first three-dimensional shape data, the first three-dimensional shape data, based on reliability of the spatial position of each of the plurality of feature points and the second three-dimensional shape data.

The present disclosure generally relates to systems and techniques for constructing three-dimensional (3D) models. Certain aspects of the present disclosure provide an apparatus for model generation. The apparatus generally includes a memory, and one or more processors coupled to the memory. The one or more processors and the memory may be configured to receive one or more images depicting an object to be modeled, determine a category associated with the object to be modeled, select a shape of a space based on the category, and generate a 3D model of the object at least in part by carving one or more points associated with the space based on the one or more images depicting the object.

A computer that includes a processor and a memory, the memory including instructions executable by the processor can activate a plurality of light sources in an alternating pattern at an illumination frequency and acquire image data depicting light reflected from an object at a timing corresponding to the alternating pattern at the illumination frequency. In response to variations in the light impinging upon the object from the alternating pattern, an object contour can be determined based on estimating one or more object surface normals based on photometric stereo. One or more of object identity and object depth can be determined based on combining the object contour with the image data.