Abstract: A surface electromyography signal-torque matching method based on multi-segmentation parallel CNN model (MSP-CNN model), step 1: collecting torque signals and surface electromyography (sEMG) signals when tightening a bolt; step 2: dividing a range of a transducer by at least two granularities, generating a plurality of torque sub-ranges corresponding to the at least two granularities and labeling the plurality of torque sub-ranges with torque labels; step 3: generating sEMG graphs of the sEMG signals in each time window; step 4: determining the torque labels of each time window under each of the at least two granularities according to the torque sub-ranges that average values of torques fall in; step 5: establishing a sample set; step 6: building a MSP-CNN model, and training parallel independent CNN models with sample datasets; and step 7: inputting the sEMG signals of the operator during assembly into trained MSP-CNN model and identifying assembly torques.

Abstract: Computer vision and image characteristic search is described. The described system leverages visual search techniques by determining visual characteristics of objects depicted in images and comparing the determined characteristics to visual characteristics of other images, e.g., to identify similar visual characteristics in the other images. In some aspects, the described system performs searches that leverage a digital image as part of a search query to locate digital content of interest. In some aspects, the described system surfaces multiple user interface instrumentalities that include images of patterns, textures, or materials and that are selectable to initiate a visual search of digital content having a similar pattern, texture, or material. The described aspects also include pattern-based authentication in which the system determines authenticity of an item in an image based on a similarity of its visual characteristics to visual characteristics of known authentic items.

Abstract: Example implementations described herein are directed to systems and methods for estimating the remaining useful life of a component or equipment through the application of models for deriving functions that can express the remaining useful life over time. In an aspect, the failure acceleration time point is determined for a given type of component, and a function is derived based on the application of models on the failure acceleration time point.

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

Abstract: Computer vision and image characteristic search is described. The described system leverages visual search techniques by determining visual characteristics of objects depicted in images and comparing the determined characteristics to visual characteristics of other images, e.g., to identify similar visual characteristics in the other images. In some aspects, the described system performs searches that leverage a digital image as part of a search query to locate digital content of interest. In some aspects, the described system surfaces multiple user interface instrumentalities that include images of patterns, textures, or materials and that are selectable to initiate a visual search of digital content having a similar pattern, texture, or material. The described aspects also include pattern-based authentication in which the system determines authenticity of an item in an image based on a similarity of its visual characteristics to visual characteristics of known authentic items.

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

Abstract: Computer vision for unsuccessful queries and iterative search is described. The described system leverages visual search techniques by determining visual characteristics of objects depicted in images and describing them, e.g., using feature vectors. In some aspects, these visual characteristics are determined for search queries that are identified as not being successful. Aggregated information describing visual characteristics of images of unsuccessful search queries is used to determine common visual characteristics and objects depicted in those images. This information can be used to inform other users about unmet needs of searching users. In some aspects, these visual characteristics are used in connection with iterative image searches where users select an initial query image and then the search results are iteratively refined.

Abstract: Camera platform techniques are described. In an implementation, a plurality of digital images and data describing times, at which, the plurality of digital images are captured is received by a computing device. Objects of clothing are recognized from the digital images by the computing device using object recognition as part of machine learning. A user schedule is also received by the computing device that describes user appointments and times, at which, the appointments are scheduled. A user profile is generated by the computing device by training a model using machine learning based on the recognized objects of clothing, times at which corresponding digital images are captured, and the user schedule. From the user profile, a recommendation is generated by processing a subsequent user schedule using the model as part of machine learning by the computing device.

Abstract: Camera platform techniques are described. In an implementation, a plurality of digital images and data describing times, at which, the plurality of digital images are captured is received by a computing device. Objects of clothing are recognized from the digital images by the computing device using object recognition as part of machine learning. A user schedule is also received by the computing device that describes user appointments and times, at which, the appointments are scheduled. A user profile is generated by the computing device by training a model using machine learning based on the recognized objects of clothing, times at which corresponding digital images are captured, and the user schedule. From the user profile, a recommendation is generated by processing a subsequent user schedule using the model as part of machine learning by the computing device.

Abstract: Example implementations involve a system for a system and method for Predictive Maintenance using Trace Norm Generative Adversarial Networks. Such example implementations can involve providing generated sensor data and real sensor data to a first network and to a second network, the first network configured to enforce trace norm minimization of the second network, the second network configured to distinguish between the generated sensor data and the real sensor data, the first network involving a subset of layers from the second network, and the second network integrated into a generative adversarial network.

Abstract: Example implementations described herein involve a system for Predictive Maintenance using Discriminant Generative Adversarial Networks, and can involve providing generated sensor data and real sensor data to a first network and to a second network, the first network configured to enforce a discriminant loss objective of the second network, the second network configured to distinguish between the generated sensor data and the real sensor data, the first network including a subset of layers from the second network, the real sensor data including pairs of real sensor data and labels, the second network integrated into a generative adversarial network (GAN); training the machine health classification model from the output of the first network using the provided generated sensor data and the real sensor data, the output of the first network including feature vectors; and deploying the machine health classification model with the first network.

Abstract: Systems, methods, and computer program products for identifying a candidate product in an electronic marketplace based on a visual comparison between candidate product image visual text content and input query image visual text content. Unlike conventional optical character recognition (OCR) based systems, embodiments automatically localize and isolate portions of a candidate product image and an input query image that each contain visual text content, and calculate a visual similarity measure between the respective portions. A trained neural network may be re-trained to more effectively find visual text content by using the localized and isolated visual text content portions as additional ground truths. The visual similarity measure serves as a visual search result score for the candidate product. Any number of images of any number of candidate products may be compared to an input query image to enable text-in-image based product searching without resorting to conventional OCR techniques.

Abstract: Techniques and systems are described that leverage computer vision as part of search to expand functionality of a computing device available to a user and increase operational computational efficiency as well as efficiency in user interaction. In a first example, user interaction with items of digital content is monitored. Computer vision techniques are used to identify digital images in the digital content, objects within the digital images, and characteristics of those objects. This information is used to assign a user to a user segment of a user population which is then used to control output of subsequent digital content to the user, e.g., recommendations, digital marketing content, and so forth.

Abstract: Camera platform techniques are described. In an implementation, a plurality of digital images and data describing times, at which, the plurality of digital images are captured is received by a computing device. Objects of clothing are recognized from the digital images by the computing device using object recognition as part of machine learning. A user schedule is also received by the computing device that describes user appointments and times, at which, the appointments are scheduled. A user profile is generated by the computing device by training a model using machine learning based on the recognized objects of clothing, times at which corresponding digital images are captured, and the user schedule. From the user profile, a recommendation is generated by processing a subsequent user schedule using the model as part of machine learning by the computing device.

Abstract: A machine is configured to generate in real time personalized online banner images for users based on data pertaining to user behavior in relation to an image of a product. For example, the machine receives a user selection indicating one or more data features associated with the user. The one or more data features include a data feature pertaining to user behavior in relation to an image of a product. The machine generates, using a machine learning algorithm, a data representation of the machine learning algorithm based on the one or more data features including the data feature pertaining to user behavior in relation to the image of the product. The data representation includes one or more data features pertaining to one or more characteristics of online banner images. The machine generates an online banner image for the user based on the data representation.

Abstract: Systems, methods, and computer program products for identifying a candidate product in an electronic marketplace based on a visual comparison between candidate product image visual text content and input query image visual text content. Unlike conventional optical character recognition (OCR) based systems, embodiments automatically localize and isolate portions of a candidate product image and an input query image that each contain visual text content, and calculate a visual similarity measure between the respective portions. A trained neural network may be re-trained to more effectively find visual text content by using the localized and isolated visual text content portions as additional ground truths. The visual similarity measure serves as a visual search result score for the candidate product. Any number of images of any number of candidate products may be compared to an input query image to enable text-in-image based product searching without resorting to conventional OCR techniques.

Abstract: Techniques and systems are described that leverage computer vision as part of search to expand functionality of a computing device available to a user and increase operational computational efficiency as well as efficiency in user interaction. In a first example, user interaction with items of digital content is monitored. Computer vision techniques are used to identify digital images in the digital content, objects within the digital images, and characteristics of those objects. This information is used to assign a user to a user segment of a user population which is then used to control output of subsequent digital content to the user, e.g., recommendations, digital marketing content, and so forth.

Abstract: Camera platform and object inventory control techniques are described. In an implementation a live feed of digital images is output in a user interface by a computing device. A user selection is received through interaction with the user interface of at least one of the digital images. An object, included within the at least one digital image, is recognized using machine learning. Metadata is then obtained that pertains to the recognized object. Augmented reality digital content is generated based at least in part of the obtained metadata. The augmented reality digital content is displayed as part of the live feed of digital images as associated with the object.

Abstract: Privacy-preserving activity monitoring systems and methods are described. In one embodiment, a plurality of sensors is configured for contact-free monitoring of at least one user state. A signal processing module communicatively coupled to the sensors is configured to receive data from the sensors. A first sensor is configured to generate a first set of quantitative data associated with a first user state. A second sensor is configured to generate a second set of quantitative data associated with a second user state. A third sensor is configured to generate a third set of quantitative data associated with a third user state. The signal processing module is configured to process the three sets of quantitative data using a machine learning module, and identify a user activity and detect a condition associated with the user, where no user-identifying information is communicated more than 100 meters to or from the signal processing module.

Abstract: Systems and methods configured to perform monitoring of one or more user activities are described. In one embodiment, a plurality of sensors is configured for contact-free monitoring of at least one bodily function. A signal processing module communicatively coupled to the sensors is configured to receive data from the sensors. A first sensor is configured to generate a first set of quantitative data associated with a user speed and a user position. A second sensor is configured to generate a second set of quantitative data associated with a user action. A third sensor is configured to generate a third set of quantitative data associated with a user movement. The signal processing module is configured to process the three sets of quantitative data using a machine learning module, and identify a user activity and detect a condition associated with the user.