Abstract: Automatic multi-objective hardware optimization for processing a deep learning network is disclosed. An example of a storage medium includes instructions for obtaining client preferences for a plurality of performance indicators for processing of a deep learning workload; generating a workload representation for the deep learning workload; providing the workload representation to machine learning processing to generate a workload executable, the workload executable including hardware mapping based on the client preferences; and applying the workload executable in processing of the deep learning workload.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

Abstract: An apparatus is provided to use NEMO search to train GNNs that can be used for mixed-precision quantization of DNNs. For example, the apparatus generates a plurality of GNNs. The apparatus further generates a plurality of new GNNs based on the plurality of GNNs. The apparatus also generates a sequential graph for a first DNN. The first DNN includes a sequence of quantizable operations, each of which includes quantizable parameters and is represented by a different node in the sequential graph. The apparatus inputs the sequential graph into the GNNs and new GNNs and evaluates outputs of the GNNs and new GNNs based on conflicting objectives of reducing precisions of the quantizable parameters of the first DNN. The apparatus then selects a GNN from the GNNs and new GNNs based on the evaluation. The GNN is to be used for reducing precisions of quantizable parameters of a second DNN.

Abstract: A system and method for resolving an ambiguity between similar objects in an image is disclosed. A three dimensional representation of a room is generated, and objects in the room are identified from an image of the room are identified. A determination is made that at least two objects are visually similar, and a position of the two objects is ambiguous. At least one question based on the determined ambiguity is programmatically generated based on information known about the room, and is phrased such that the ambiguity can be resolved by an answer to the question. Based on the answer received one of the objects is selected. At least one property of the selected object is modified based upon the selection of one of the at least two objects.

Abstract: Automatic multi-objective hardware optimization for processing a deep learning network is disclosed. An example of a storage medium includes instructions for obtaining client preferences for a plurality of performance indicators for processing of a deep learning workload; generating a workload representation for the deep learning workload; providing the workload representation to machine learning processing to generate a workload executable, the workload executable including hardware mapping based on the client preferences; and applying the workload executable in processing of the deep learning workload.

Abstract: A method, an apparatus and a program for automatic generation of secondary class annotations. The method comprises obtaining a plurality of images of an environment, each of which comprising objects in the environment. Some of the objects are annotated, while other objects are not. The method comprises aligning the plurality of images to a common coordinates system and computing a plurality of weighted images by adding weights to regions in the plurality of images that are associated with annotated objects to reduce significance of such regions. The method further comprises generating, based on the plurality of weighted images, a background model of the environment by determining for each region in the common coordinates system a statistical metric representing a visual feature of a background of the environment. The background model is then utilized to identify the non-annotated objects and adding an annotation for each identified object.

Abstract: A method can include learning a common embedding space and a set of parameters for each one of a plurality of sets of mixture models, wherein one mixture model is associated with one class of objects within a set of object categories. The method can also include adding new mixture models to the set of mixture models to support novel categories based on a set of example embedding vectors computed for each one of the novel categories. Additionally, the method includes detecting in images a plurality of boxes with associated labels and corresponding confidence scores, wherein the boxes correspond to image regions comprising objects of both known categories and the novel categories. Furthermore, the method includes, given a query image, executing an instruction based on the common embedding space and the set of mixture models, the instruction comprising identifying objects from both categories in the query image.

Abstract: Embodiments of the present disclosure include training a model using a plurality of pairs of feature vectors related to a first class. Embodiments include providing a sample feature vector related to a second class as an input to the model. Embodiments include receiving at least one synthesized feature vector as an output from the model. Embodiments include training a classifier to recognize the second class using a training data set comprising the sample feature vector related to the second class and the at least one synthesized feature vector. Embodiments include providing a query feature vector as an input to the classifier. Embodiments include receiving output from the classifier that identifies the query feature vector as being related to the second class, wherein the output is used to perform an action.

Abstract: A method can include learning a common embedding space and a set of parameters for each one of a plurality of sets of mixture models, wherein one mixture model is associated with one class of objects within a set of object categories. The method can also include adding new mixture models to the set of mixture models to support novel categories based on a set of example embedding vectors computed for each one of the novel categories. Additionally, the method includes detecting in images a plurality of boxes with associated labels and corresponding confidence scores, wherein the boxes correspond to image regions comprising objects of both known categories and the novel categories. Furthermore, the method includes, given a query image, executing an instruction based on the common embedding space and the set of mixture models, the instruction comprising identifying objects from both categories in the query image.

Abstract: Embodiments of the present disclosure include training a model using a plurality of pairs of feature vectors related to a first class. Embodiments include providing a sample feature vector related to a second class as an input to the model. Embodiments include receiving at least one synthesized feature vector as an output from the model. Embodiments include training a classifier to recognize the second class using a training data set comprising the sample feature vector related to the second class and the at least one synthesized feature vector. Embodiments include providing a query feature vector as an input to the classifier. Embodiments include receiving output from the classifier that identifies the query feature vector as being related to the second class, wherein the output is used to perform an action.

Abstract: A method, an apparatus and a program for automatic generation of secondary class annotations. The method comprises obtaining a plurality of images of an environment, each of which comprising objects in the environment. Some of the objects are annotated, while other objects are not. The method comprises aligning the plurality of images to a common coordinates system and computing a plurality of weighted images by adding weights to regions in the plurality of images that are associated with annotated objects to reduce significance of such regions. The method further comprises generating, based on the plurality of weighted images, a background model of the environment by determining for each region in the common coordinates system a statistical metric representing a visual feature of a background of the environment. The background model is then utilized to identify the non-annotated objects and adding an annotation for each identified object.

Abstract: Embodiments of the present invention may provide automated techniques for quickly and easily installing a different model for image recognition on each of numerous devices without doing manual configuration of each and every device. For example, in an embodiment, a computer-implemented method for configuring devices may comprise capturing data at a device, transmitting the captured data to a remote system, receiving configuration data for the device, wherein the configuration data has been generated by processing the captured data at the remote system using a plurality of different machine learning techniques and generating the configuration data based on at least one of the plurality of different machine learning techniques, and configuring the device using the received configuration data.

Abstract: A system and method for resolving an ambiguity between similar objects in an image is disclosed. A three dimensional representation of a room is generated, and objects in the room are identified from an image of the room are identified. A determination is made that at least two objects are visually similar, and a position of the two objects is ambiguous. At least one question based on the determined ambiguity is programmatically generated based on information known about the room, and is phrased such that the ambiguity can be resolved by an answer to the question. Based on the answer received one of the objects is selected. At least one property of the selected object is modified based upon the selection of one of the at least two objects.

Abstract: Embodiments of the present invention may provide automated techniques for checking store shelves for compliance with planograms that can handle unknown arrangements in a uniform way, with little user involvement, and with relatively low processing complexity. For example, a computer-implemented method for determining compliance with a planogram may comprise receiving at least one image of a plurality of shelves containing objects, receiving at least one planogram representing desired positions of the objects on the shelves, wherein each row of the planogram corresponds to one shelf, adjusting positions of objects in the planogram to reflect a size of the shelves, determining a tolerance for inaccuracy in object position in the image using object dimension data, detecting object positions in the image, aligning each row of the planogram individually to the detected object positions, and comparing the detected object positions with the adjusted planogram positions to generate compliance information.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

Abstract: There is provided a method of identifying objects in an image, comprising: extracting query descriptors from the image, comparing each query descriptor with training descriptors for identifying matching training descriptors, each training descriptor is associated with a reference object identifier and with relative location data (distance and direction from a center point of a reference object indicated by the reference object identifier), computing object-regions of the digital image by clustering the query descriptors having common center points defined by the matching training descriptors, each object-region approximately bounding one target object and associated with a center point and a scale relative to a reference object size, wherein the object-regions are computed independently of the identifier of the reference object associated with the object-regions, wherein members of each cluster point toward a common center point, and classifying the target object of each object-region according to the referen

Abstract: Embodiments of the present invention may provide automated techniques for quickly and easily installing a different model for image recognition on each of numerous devices without doing manual configuration of each and every device. For example, in an embodiment, a computer-implemented method for configuring devices may comprise capturing data at a device, transmitting the captured data to a remote system, receiving configuration data for the device, wherein the configuration data has been generated by processing the captured data at the remote system using a plurality of different machine learning techniques and generating the configuration data based on at least one of the plurality of different machine learning techniques, and configuring the device using the received configuration data.