In an example, a method includes, by one or more processors, receiving a greyscale image having a plurality of pixels, each pixel being associated with a grey level, and the greyscale image having a first number of grey levels. An order of the pixels may be determined based on the grey level. A second number of grey levels may be determined, wherein the second number of grey levels is greater than the first number, and an indication of a target number of pixels per grey level of the second number of grey levels may be further be determined. Taking the pixels in order, and based on the target number of pixels per grey level, a new grey level may be allocated to each pixel to provide the second number of grey levels. The new grey levels may be converted to a threshold of a threshold halftone screen.

A method may include generating views materializing tensors generated by a convolutional neural network operating on an image. Determining the outputs of the convolutional neural network operating on the image with a patch occluding various portions of the image. The outputs being determined by generating queries on the views that performs, based at least on the changes associated with occluding different portions of the image, partial re-computations of the views. A heatmap may be generated based on the outputs of the convolutional neural network. The heatmap may indicate the quantities to which the different portions of the image contribute to the output of the convolutional neural network operating on the image. Related systems and articles of manufacture, including computer program products, are also provided.

An electronic device is provided. The electronic device includes a memory, an image sensor including light receiving elements each including at least two sub light receiving elements, and an image signal processor. The image signal processor is configured to obtain images corresponding to light from outside by using the image sensor, the images including at least a raw image, a first sub image, and a second sub image, the first sub image being an image corresponding to light detected by at least one first sub light, the second sub image being an image corresponding to light detected by at least one second sub light, identify a luminance ratio between the first sub image and the second sub image, identify a defect in the raw image, based on the luminance ratio, and perform a function corresponding to a type of the defect.

Disclosed is a method for performing image convolution by considering a hierarchical relationship of hyperbolic feature vectors in a hyperbolic space. According to an embodiment of the present disclosure, an image convolution method in a hyperbolic space includes steps of embedding an image feature vector on a Euclidean space into a hyperbolic feature vector on a hyperbolic space, allocating a hierarchical weight on the hyperbolic feature vector based on a hierarchical property of the hyperbolic feature vector, and convolutioning the hyperbolic feature vector by applying the hierarchical weight.

A system and method of detecting an image of a template object in a captured image may include comparing, by a processor, an image model of an imaged template object to multiple locations, rotations, and scales in the captured image. The image model may be defined by multiple model base point sets derived from contours of the imaged template object, where each model base point set inclusive of a plurality of model base points that are positioned at corresponding locations associated with distinctive features of the imaged template object. Each corresponding model base point of the model base point sets may (i) be associated with respective layers and (ii) have an associated gradient vector. A determination may be made as to whether and where the image of the object described by the image model is located in the captured image.

A visual sensor having a light sensitive element and a processor, the processor being adapted to recognize whether a cap is on or off the light sensitive element by recognizing a unique identification pattern coded into the light sensitive element.

This disclosure describes systems and methods for training a neural network with a training data set including data items labeled at different granularities. During training, each item within the training data set can be fed through the neural network. For items with labels of a higher granularity, weights of the network can be adjusted based on a comparison between the output of the network and the label of the item. For items with labels of a lower granularity, an output of the network can be fed through a conversion function that convers the output from the higher granularity to the lower granularity. The weights of the network can then be adjusted based on a comparison between the converted output and the label of the item.

A computing system including a processing circuit in communication with a camera having a field of view. The processing circuit is configured to perform operations related to detecting, identifying, and retrieving objects disposed amongst a plurality of objects. The processing circuit may be configured to perform operations related to object recognition template generation, feature generation, hypothesis generation, hypothesis refinement, and hypothesis validation.

Systems and methods for image-based perception. The methods comprise: obtaining, by a computing device, images captured by a plurality of cameras with overlapping fields of view; generating, by the computing device, spatial feature maps indicating locations of features in the images; defining, by the computing device, predicted cuboids at each location of an object in the images based on the spatial feature maps; and assigning, by the computing device, at least two cuboids of said predicted cuboids to a given object when predictions from images captured by separate cameras of the plurality of cameras should be associated with a same detected object.

In some implementations, a device may receive data identifying products and encoded data identifying smart tags of the products. The device may map the data and the encoded data to generate encoded product data. The device may receive encoded data provided by smart tags of products received by a store. The device may receive images of the products. The device may compare the encoded data and the encoded product data to identify a set of the products received by the store. The device may correlate the images with the set of the products. The device may process the correlated data to identify locations of the set of the products in the store. The device may generate an instruction to relocate a product to a new location and may provide the instruction to a device, associated with the store, to cause the product to be relocated to the new location.