Abstract: In one embodiment, a method includes receiving an input video comprising a plurality of image frames including an object of interest. Based on the plurality of image frames, a motion associated with the object of interest is determined, and the plurality of image frames are classified using a machine-learning model to identify one of the plurality of image frames that indicates a moment of perception of the determined motion.

Abstract: Systems and methods are disclosed for training neural networks using labels for training data that are dynamically refined using neural networks and using these trained neural networks to perform detection and/or classification of one or more objects appearing in an image. Particular embodiments may generate a set of crops of images from a corpus of images, then apply a first neural network to the set of crops to obtain a set of respective outputs. A second neural network may then be trained using the set of crops as training examples. The set of respective outputs may be applied as labels for the set of crops.

Abstract: In one embodiment, a method includes receiving an input video comprising a plurality of image frames including an object of interest. Based on the plurality of image frames, a motion associated with the object of interest is determined, and the plurality of image frames are classified using a machine-learning model to identify one of the plurality of image frames that indicates detection of the determined motion.

Abstract: In one embodiment, a method includes receiving an input vector corresponding to a query at a neural network model comprising a plurality of layers, wherein the plurality of layers comprise a last layer associated with a mapping matrix, generating a binary matrix based on the mapping matrix, an identity matrix, and one or more Gaussian vectors, generating an integer vector based on the binary matrix and a binary vector associated with the input vector, identifying a plurality of indices corresponding to a plurality of top values of the integer vector for the integer vector, generating an output vector based on the input vector and a plurality of rows of the mapping matrix, wherein the plurality of rows is associated with the plurality of identified indices, respectively, and determining the query is associated with one or more classes based on the output vector.

Abstract: Systems and processes for training and compressing a convolutional neural network model include the use of quantization and layer fusion. Quantized training data is passed through a convolutional layer of a neural network model to generate convolutional results during a first iteration of training the neural network model. The convolutional results are passed through a batch normalization layer of the neural network model to update normalization parameters of the batch normalization layer. The convolutional layer is fused with the batch normalization layer to generate a first fused layer and the fused parameters of the fused layer are quantized. The quantized training data is passed through the fused layer using the quantized fused parameters to generate output data, which may be quantized for a subsequent layer in the training iteration.

Abstract: In one embodiment, a method includes providing, to a client system of a user, a user interface for display. The user interface may include a first set of options for selecting an artificial intelligence (AI) task for integrating into a user application, a second set of options for selecting one or more devices on which the user wants to deploy the selected AI task, and a third set of options for selecting one or more performance constraints specific to the selected devices. User specifications may be received based on user selections in the first, second, and third sets of options. A custom AI model may be generated based on the user specifications and sent to the client system of the user for integrating into the user application. The custom AI model once integrated may enable the user application to perform the selected AI task on the selected devices.

Abstract: Systems and methods are disclosed for lookup-based convolutional neural networks. For example, methods may include applying a convolutional neural network to image data based on an image to obtain an output, in which a layer of the convolutional network includes filters with weights that are stored as a dictionary (D) of channel weight vectors, a respective lookup index tensor (I) that indexes the dictionary, and a respective lookup coefficient tensor (C), and in which applying the convolutional neural network includes: convolving the channel weight vectors of the dictionary (D) with an input tensor based on the image to obtain an input dictionary (S), and combining entries of the input dictionary (S) that are indexed with indices from the respective lookup index tensor (I) and multiplied with corresponding coefficients from the respective lookup coefficient tensor (C); and storing, displaying, or transmitting data based on the output of the convolutional neural network.

Abstract: In one embodiment, a method includes providing, to a client system of a user, a user interface for display. The user interface may include a first set of options for selecting an artificial intelligence (AI) task for integrating into a user application, a second set of options for selecting one or more devices on which the user wants to deploy the selected AI task, and a third set of options for selecting one or more performance constraints specific to the selected devices. User specifications may be received based on user selections in the first, second, and third sets of options. A custom AI model may be generated based on the user specifications and sent to the client system of the user for integrating into the user application. The custom AI model once integrated may enable the user application to perform the selected AI task on the selected devices.

Abstract: In one embodiment, a method includes receiving an input video comprising a plurality of image frames including an object of interest. Based on the plurality of image frames, a motion associated with the object of interest is determined, and the plurality of image frames are classified using a machine-learning model to identify one of the plurality of image frames that indicates detection of the determined motion.

Abstract: Systems and methods are disclosed for training neural networks using labels for training data that are dynamically refined using neural networks and using these trained neural networks to perform detection and/or classification of one or more objects appearing in an image. Particular embodiments may generate a set of crops of images from a corpus of images, then apply a first neural network to the set of crops to obtain a set of respective outputs. A second neural network may then be trained using the set of crops as training examples. The set of respective outputs may be applied as labels for the set of crops.

Abstract: Systems and methods are disclosed for training neural networks using labels for training data that are dynamically refined using neural networks and using these trained neural networks to perform detection and/or classification of one or more objects appearing in an image. Particular embodiments may generate a set of crops of images from a corpus of images, then apply a first neural network to the set of crops to obtain a set of respective outputs. A second neural network may then be trained using the set of crops as training examples. The set of respective outputs may be applied as labels for the set of crops.

Abstract: Aspects of the subject technology relate to machine learning based object recognition using pixel difference information. A difference image generated by subtraction of a current image from one or more previous images can be provided, as input, to a machine-learning engine. The machine-learning may output a detected object or a detected action based, at least in part, on the difference image. In this way, temporal information about the object can be provided to, and used by, a machine-learning model that is structured to accept image input.

Abstract: In one embodiment, a method includes receiving a machine-learning model trained to detect a specified motion using multiple videos, wherein each video has at least one frame labeled as a moment of perception of the specified motion, identifying an object-of-interest depicted in an input video, detecting a motion of the object-of-interest, determining that the detected motion is the specified motion, and classifying one of the frames of the input video as the moment of perception of the specified motion.

Abstract: In one embodiment, a method includes receiving a machine-learning model trained to detect a specified motion using multiple videos, wherein each video has at least one frame labeled as a moment of perception of the specified motion, identifying an object-of-interest depicted in an input video, detecting a motion of the object-of-interest, determining that the detected motion is the specified motion, and classifying one of the frames of the input video as the moment of perception of the specified motion.

Abstract: In one embodiment, a method includes receiving an input vector corresponding to a query at a neural network model comprising a plurality of layers, wherein the plurality of layers comprise a last layer associated with a mapping matrix, generating a binary matrix based on the mapping matrix, an identity matrix, and one or more Gaussian vectors, generating an integer vector based on the binary matrix and a binary vector associated with the input vector, identifying a plurality of indices corresponding to a plurality of top values of the integer vector for the integer vector, generating an output vector based on the input vector and a plurality of rows of the mapping matrix, wherein the plurality of rows is associated with the plurality of identified indices, respectively, and determining the query is associated with one or more classes based on the output vector.

Abstract: Systems and methods are disclosed for lookup-based convolutional neural networks. For example, methods may include applying a convolutional neural network to image data based on an image to obtain an output, in which a layer of the convolutional network includes filters with weights that are stored as a dictionary (D) of channel weight vectors, a respective lookup index tensor (I) that indexes the dictionary, and a respective lookup coefficient tensor (C), and in which applying the convolutional neural network includes: convolving the channel weight vectors of the dictionary (D) with an input tensor based on the image to obtain an input dictionary (S), and combining entries of the input dictionary (S) that are indexed with indices from the respective lookup index tensor (I) and multiplied with corresponding coefficients from the respective lookup coefficient tensor (C); and storing, displaying, or transmitting data based on the output of the convolutional neural network.

Abstract: Systems and processes for training and compressing a convolutional neural network model include the use of quantization and layer fusion. Quantized training data is passed through a convolutional layer of a neural network model to generate convolutional results during a first iteration of training the neural network model. The convolutional results are passed through a batch normalization layer of the neural network model to update normalization parameters of the batch normalization layer. The convolutional layer is fused with the batch normalization layer to generate a first fused layer and the fused parameters of the fused layer are quantized. The quantized training data is passed through the fused layer using the quantized fused parameters to generate output data, which may be quantized for a subsequent layer in the training iteration.

Abstract: Systems and methods are disclosed for lookup-based convolutional neural networks. For example, methods may include applying a convolutional neural network to image data based on an image to obtain an output, in which a layer of the convolutional network includes filters with weights that are stored as a dictionary (D) of channel weight vectors, a respective lookup index tensor (I) that indexes the dictionary, and a respective lookup coefficient tensor (C), and in which applying the convolutional neural network includes: convolving the channel weight vectors of the dictionary (D) with an input tensor based on the image to obtain an input dictionary (S), and combining entries of the input dictionary (S) that are indexed with indices from the respective lookup index tensor (I) and multiplied with corresponding coefficients from the respective lookup coefficient tensor (C); and storing, displaying, or transmitting data based on the output of the convolutional neural network.

Abstract: In one embodiment, a method includes providing, to a client system of a user, a user interface for display. The user interface may include a first set of options for selecting an artificial intelligence (AI) task for integrating into a user application, a second set of options for selecting one or more devices on which the user wants to deploy the selected AI task, and a third set of options for selecting one or more performance constraints specific to the selected devices. User specifications may be received based on user selections in the first, second, and third sets of options. A custom AI model may be generated based on the user specifications and sent to the client system of the user for integrating into the user application. The custom AI model once integrated may enable the user application to perform the selected AI task on the selected devices.

Abstract: Systems and methods are disclosed for training neural networks using labels for training data that are dynamically refined using neural networks and using these trained neural networks to perform detection and/or classification of one or more objects appearing in an image. Particular embodiments may generate a set of crops of images from a corpus of images, then apply a first neural network to the set of crops to obtain a set of respective outputs. A second neural network may then be trained using the set of crops as training examples. The set of respective outputs may be applied as labels for the set of crops.