A computer-implemented method for surface modeling includes: receiving one or more polarization raw frames of a surface of a physical object, the polarization raw frames being captured with a polarizing filter at different linear polarization angles; extracting one or more first tensors in one or more polarization representation spaces from the polarization raw frames; and detecting a surface characteristic of the surface of the physical object based on the one or more first tensors in the one or more polarization representation spaces.

A computer-implemented method for surface modeling includes: receiving one or more polarization raw frames of a surface of a physical object, the polarization raw frames being captured with a polarizing filter at different linear polarization angles; extracting one or more first tensors in one or more polarization representation spaces from the polarization raw frames; and detecting a surface characteristic of the surface of the physical object based on the one or more first tensors in the one or more polarization representation spaces.

A non-transitory computer-readable medium encoded with a computer-readable program which, when executed by a processor, will cause a computer to execute a method of authenticating a 3-D object with a 2-D camera, the method including building a pre-determined database. The method additionally includes registering the 3-D object to a storage unit of a device comprising the 2-D camera, thereby creating a registered 3-D model of the 3-D object. Additionally, the method includes authenticating a test 3-D object by comparing the test 3-D object to the registered 3-D model.

Systems and methods are disclosed for determining textual transcription from minute facial skin movements. In one implementation, a system may include at least one coherent light source, at least one sensor configured to receive light reflections from the at least one coherent light source; and a processor configured to control the at least one coherent light source to illuminate a region of a face of a user. The processor may receive from the at least one sensor, reflection signals indicative of coherent light reflected from the face in a time interval. The reflection signals may be analyzed to determine minute facial skin movements in the time interval. Then, based on the determined minute facial skin movements in the time interval, the processor may determine a sequence of words associated with the minute facial skin movements, and output a textual transcription corresponding with the determined sequence of words.

Examples of the disclosure describe systems and methods for identifying, quantifying, and/or characterizing plant stomata. In an example method, a first set of two or more images of a plant leaf representing two or more focal distances is captured via an optical sensor. A reference focal distance is determined based on the first set of images. A second set of two or more images of the plant leaf is captured via the optical sensor, including at least one image captured at a focal distance less than the reference focal distance, and at least one image captured at a focal distance greater than the reference focal distance. A composite image is generated based on the second set of images. The composite image is provided to a trainable feature detector in order to determine a number, density, and/or distribution of stomata in the composite image.

A system for collecting data for training a computer vision model for shape estimation includes: an imaging system configured to capture one or more images; and a processing system including a processor and memory storing instructions that, when executed by the processor, cause the processor to: receive one or more input images from the imaging system; estimate a pose of an object depicted in the one or more images; render a shape estimate from a 3-D model of the object posed in accordance with the pose of the object; and generate a data point of a training dataset, the data point including one or more images based on the one or more input images and a label corresponding to the one or more images, the label including the shape estimate.

A method for performing region-of-interest (ROI)-based depth detection with aid of a pattern-adjustable projector and associated apparatus are provided. The method includes: utilizing a first camera to capture a first image, wherein the first image includes image contents indicating one or more objects; utilizing an image processing circuit to determine a ROI of the first image according to the image contents of the first image; utilizing the image processing circuit to perform projection region selection to determine a selected projection region corresponding to the ROI among multiple predetermined projection regions, wherein the selected projection region is selected from the multiple predetermined projection regions according to the ROI; utilizing the pattern-adjustable projector to project a predetermined pattern according to the selected projection region, for performing depth detection; utilizing a second camera to capture a second image; and performing the depth detection according to the second image to generate a depth map.

A depth camera assembly (DCA) has a light source assembly, a mask, a camera assembly, and a controller. The light source assembly includes at least one light source. The mask is configured to generate an interference pattern that is projected into a target area. The mask has two openings configured to pass through light emitted by the at least one light source, and the light passed through the two openings forms an interference pattern across the target area. The interference pattern has a phase based on a position of the light source. The camera assembly is configured to capture images of a portion of the target area that includes the interference pattern. The controller is configured to determine depth information for the portion of the target area based on the captured images.

A display system for a medical suite comprises a scanning device configured to capture scanning data in the medical suite. At least one display configured to display information in an operating region of the medical suite. A controller is in communication with the scanning device and the display. The controller is configured to control the scanning device to capture identifying information of a patient. Based on the identifying information, the controller is configured to authenticate an identity of the patient. Based on the identity, the controller is configured to access a patient record for the patient. The controller is further configured to control the at least one display to display information based on the patient record.

During a training phase, a first machine learning system is trained using actual data, such as multimodal images of a hand, to generate synthetic image data. During training, the first system determines latent vector spaces associated with identity, appearance, and so forth. During a generation phase, latent vectors from the latent vector spaces are generated and used as input to the first machine learning system to generate candidate synthetic image data. The candidate image data is assessed to determine suitability for inclusion into a set of synthetic image data that may be used for subsequent use in training a second machine learning system to recognize an identity of a hand presented by a user. For example, the candidate synthetic image data is compared to previously generated synthetic image data to avoid duplicative synthetic identities. The second machine learning system is then trained using the approved candidate synthetic image data.