Abstract: This disclosure relates generally to the field of image processing, and, more particularly, to a method and system for Multimodal Image Super-Resolution (MISR) using convolutional dictionary learning. Existing sparse representation learning based techniques for MISR have certain limitations which impact quality of the reconstructed image. The present disclosure performs MISR using convolutional dictionaries which are translation invariant. Low-resolution image of the target modality and high-resolution image of the guidance modality are modelled using their respective convolutional dictionaries and associated coefficients. Additionally, two coupling convolutional dictionaries are learned to model the relationship between them and synthesize the high-resolution image of the target modality more efficiently.

Abstract: Existing Convolutional Dictionary Learning (CDL) based machine fault classification do not utilize label information while learning the dictionary, hence the representation learned are not class-discriminative. Method and system disclosed herein provide a label-consistent convolutional dictionary learning approach for machine fault classification. The approach involves generating a training data for a classifier, wherein coefficients forming a plurality of class-discriminative features form the training data. The training data is then used to train a classifier, which is then used to perform machine fault classification for a given test data.

Abstract: This disclosure relates generally to a field of industrial machine inspection, and, more particularly, to method and system for acoustic based industrial machine inspection using Delay-and-Sum beamforming (DAS-BF) and dictionary learning (DL). The disclosed method presents a two-stage approach for anomaly detection using a multi-channel acoustic mixed signal. In the first stage, separation of a plurality of acoustic signals corresponding to the spatially distributed acoustic sources is performed at a coarser level by using the DAS-BF. Subsequently, dictionaries pre-trained using the plurality of acoustic signals of the individual source machines are utilized for generating a plurality of separated acoustic source signals. The generated plurality of separated acoustic source signals are analyzed for the anomaly detection by comparing them with a corresponding normal machine sound template.

Abstract: A method includes obtaining, using at least one processor, audio/video content. The method also includes processing, using the at least one processor, the audio/video content with a trained attention-based machine learning model to classify the audio/video content. Processing the audio/video content includes, using the trained attention-based machine learning model, generating a global representation of the audio/video content based on the audio/video content, generating a local representation of the audio/video content based on different portions of the audio/video content, and combining the global representation of the audio/video content and the local representation of the audio/video content to generate an output representation of the audio/video content. The audio/video content is classified based on the output representation.

Abstract: In industrial inspection scenarios, early detection of machine malfunction is extremely essential as it helps in preventing any significant damage and the associated economic losses. Embodiments herein provide a method and system for an acoustic based anomaly detection in industrial machines using a beamforming and a sequential transform learning. Herein, the system employs two-stage multi-channel source separation technique that uses the well-known delay and sum beamforming followed by a recent data-driven sequential transform learning (STL) approach to obtain clean sources. The STL is a solution to linear state-space model where operators/matrices are learnt from data and is used here to model the dynamics of time-varying source signals for source separation. Subsequently, a reference template matching is employed on each separated source to detect an anomaly.

Abstract: This disclosure relates to a method and system for multi-sensor fusion in the presence of missing and noisy labels. Prior methods for multi-sensor fusion do not estimate and correct labels for learning effective models in semi-supervised learning methods. Embodiments of the present disclosure provides a method for learning robust sensor-specific autoencoder based fusion model by utilizing a graph structure to perform label propagation and correction. In the disclosed Graph regularized AutoFuse (GAF) method latent representation for each sensor is learnt using the sensor-specific autoencoders. Further these latent representations are combined and fed to a classifier for multi-class classification. The disclosure presents a joint optimization formulation for multi-sensor fusion where label propagation and correction, sensor-specific learning and classification are executed together.

Abstract: A method includes obtaining, using at least one processor, audio/video content. The method also includes processing, using the at least one processor, the audio/video content with a trained attention-based machine learning model to classify the audio/video content. Processing the audio/video content includes, using the trained attention-based machine learning model, generating a global representation of the audio/video content based on the audio/video content, generating a local representation of the audio/video content based on different portions of the audio/video content, and combining the global representation of the audio/video content and the local representation of the audio/video content to generate an output representation of the audio/video content. The audio/video content is classified based on the output representation.

Abstract: Multi-sensor fusion is a technology which effectively utilizes the data from multiple sensors so as to portray a unified picture with improved information and offers significant advantages over existing single sensor-based techniques. This disclosure relates to a method and system for a multi-label classification using a two-stage autoencoder. Herein, the system employs autoencoder based architectures, where either raw sensor data or hand-crafted features extracted from each sensor are used to learn sensor-specific autoencoders. The corresponding latent representations from a plurality of sensors are combined to learn a fusing autoencoder. The latent representation of the fusing autoencoder is used to learn a label consistent classifier for multi-class classification. Further, a joint optimization technique is presented for learning the autoencoders and classifier weights together.

Abstract: A method includes accessing video data that includes at least two different modalities. The method also includes using a convolutional neural network layer to incorporate temporal coherence into a machine learning model architecture configured to process the video data. The method further includes learning dependency among the at least two different modalities in an attention space of the machine learning model architecture. In addition, the method includes predicting one or more correlations among the at least two different modalities.

Abstract: The disclosure herein generally relates to image processing, and, more particularly, to a method and system for impurity detection using multi-modal image processing. This system uses a combination of polarization data, and at least one of a depth data and an RGB image data to perform the impurity material detection. The system uses a graph fusion based approach while processing the captured images to detect presence of the impurity material, and accordingly alert the user.

Abstract: A method implemented by one or more computing systems includes accessing a set of content objects, in which each content object of the set of content objects is pre-labeled with concepts of a plurality of concepts organized according to a hierarchical relationship. The method further includes training, by a machine-learning model, a classification model for classifying content objects within the set of content objects. Training the classification model includes determining, for each object, a plurality of classification values corresponding to the plurality of concepts, calculating a loss for each of the plurality of classification values based on the pre-labeled concepts associated with the object, utilizing a hierarchical constraint loss function to calculate a maximum loss based on the calculated loss for each of the plurality of classification values, and updating the classification model based on the hierarchical constraint loss function until the maximum loss satisfies a predetermined criterion.

Abstract: The disclosure herein generally relates to image processing, and, more particularly, to a method and system for impurity detection using multi-modal image processing. This system uses a combination of polarization data, and at least one of a depth data and an RGB image data to perform the impurity material detection. The system uses a graph fusion based approach while processing the captured images to detect presence of the impurity material, and accordingly alert the user.