This invention provides a system and method for finding patterns in images that incorporates neural net classifiers. A pattern finding tool is coupled with a classifier that can be run before or after the tool to have labeled pattern results with sub-pixel accuracy. In the case of a pattern finding tool that can detect multiple templates, its performance is improved when a neural net classifier informs the pattern finding tool to work only on a subset of the originally trained templates. Similarly, in the case of a pattern finding tool that initially detects a pattern, a neural network classifier can then determine whether it has found the correct pattern. The neural network can also reconstruct/clean-up an imaged shape, and/or to eliminate pixels less relevant to the shape of interest, therefore reducing the search time, as well significantly increasing the chance of lock on the correct shapes.

Systems and methods for automated detection of changes in extent of structures using imagery are disclosed, including a non-transitory computer readable medium storing computer executable code that when executed by a processor cause the processor to: align, with an image classifier model, an outline of a structure at a first instance of time to pixels within an image depicting the structure captured at a second instance of time; assess a degree of alignment between the outline and the pixels depicting the structure, so as to classify similarities between the structure depicted within the pixels of the image and the outline using a machine learning model to generate an alignment confidence score; and determine an existence of a change in the structure based upon the alignment confidence score indicating a level of confidence below a predetermined threshold level of confidence that the outline and the pixels within the image are aligned.

Disclosed herein are systems and methods for determining whether an unknown product matches a scanned barcode during a checkout process. An edge computing device or other computer system can receive, from an overhead camera at a checkout lane, image data of an unknown product that is placed on a flatbed scanning area, identify candidate product identifications for the unknown product based on applying a classification model and/or product identification models to the image data, and determine based on the candidate product identifications, whether the unknown product matches a product associated with a barcode that is scanned at a POS terminal in the checkout lane. The classification model can be used to determine n-dimensional space feature values for the unknown product and determine which product the unknown product likely matches. The product identification models can be used to determine whether the unknown product is one of the products that are modeled.

Systems and methods herein describe a media capture system that receives sensor data from biometric sensors coupled to a head-wearable apparatus, detects a trigger event corresponding to a user of the head-wearable apparatus based on the sensor data, captures images using a camera coupled to the head-wearable apparatus, and transmits the captured images to a client device.

Disclosed are computer-readable devices, systems and methods for generating a model of a clothed body. The method includes generating a model of an unclothed human body, the model capturing a shape or a pose of the unclothed human body, determining two-dimensional contours associated with the model, and computing deformations by aligning a contour of a clothed human body with a contour of the unclothed human body. Based on the two-dimensional contours and the deformations, the method includes generating a first two-dimensional model of the unclothed human body, the first two-dimensional model factoring the deformations of the unclothed human body into one or more of a shape variation component, a viewpoint change, and a pose variation and learning an eigen-clothing model using principal component analysis applied to the deformations, wherein the eigen-clothing model classifies different types of clothing, to yield a second two-dimensional model of a clothed human body.

There is provided a method for generating a personalized Head Related Transfer Function (HRTF). The method can include capturing an image of an ear using a portable device, auto-scaling the captured image to determine physical geometries of the ear and obtaining a personalized HRTF based on the determined physical geometries of the ear.

A device may identify, in a first frame of a video feed captured by a camera and using a first computer vision technique, a first subject based on a plurality of reference points of the first subject. The device may determine whether the first subject is merged with a second subject in a second frame of the video feed. The device may selectively identify the first subject in the second frame using the first computer vision technique, or using a second computer vision technique, based on whether the first subject is merged with the second subject in the second frame, wherein the second computer vision technique is based on a shape context of the first subject. The device may determine log information based on identifying the first subject in the first frame and the second frame. The device may store or provide the log information.

The present disclosure describes multi-stage image editing techniques to improve detail and accuracy in edited images. An input image including a target region to be edited and an edit parameter specifying a modification to the target region are received. A parsing map of the input image is generated. A latent representation of the parsing map is generated. An edit is applied to the latent representation of the parsing map based on the edit parameter. The edited latent representation is input to a neural network to generate a modified parsing map including the target region with a shape change according to the edit parameter. Based on the input image and the modified parsing map, a masked image corresponding to the shape change is generated. Based on the masked image, a neural network is used to generate an edited image with the modification to the target region.

Generating visual data by defining a first action into a first set of objects and corresponding first set of motions, and defining a second action into a second set of objects and corresponding second set of motions. A relationship is then determined for the second action to the first action in terms of relationships between corresponding constituent objects and motions. Objects and motions are detected from visual data of first action. Visual data is composed for the second action from the data by transforming the constituent objects and motions detected in first action based on the corresponding determined relationships.

There is provided a method for generating a personalized Head Related Transfer Function (HRTF). The method can include capturing an image of an ear using a portable device, auto-scaling the captured image to determine physical geometries of the ear and obtaining a personalized HRTF based on the determined physical geometries of the ear.