Extracting multiscale contextual information and higher-order correlations among skeleton sequences using Graph Convolutional Networks (GCNs) alone is inadequate for effective action classification. Hypergraph convolution addresses the above issues but cannot harness the long-range dependencies. The transformer proves to be effective in capturing these dependencies and making complex contextual features accessible. We propose an Autoregressive Adaptive HyperGraph Transformer (AutoregAd-HGformer) model for in-phase (autoregressive and discrete) and out-phase (adaptive) hypergraph generation. The vector quantized in-phase hypergraph equipped with powerful autoregressive learned priors produces a more robust and informative representation suitable for hyperedge formation. The out-phase hypergraph generator provides a model-agnostic hyperedge learning technique to align the attributes with input skeleton embedding. The hybrid (supervised and unsupervised) learning in AutoregAd-HGformer explores the action-dependent feature along spatial, temporal, and channel dimensions. The extensive experimental results and ablation study indicate the superiority of our model over state-of-the-art hypergraph architectures on the NTU RGB+D, NTU RGB+D 120, and NW-UCLA datasets. Find the code Here.

Transformer-based models have revolutionized the field of image super-resolution (SR) by harnessing their inherent ability to capture complex contextual features. The overlapping rectangular shifted window technique used in transformer architecture nowadays is a common practice in super-resolution models to improve the quality and robustness of image upscaling. However, it suffers from distortion at the boundaries and has limited unique shifting modes. To overcome these weaknesses, we propose a non-overlapping triangular window technique that synchronously works with the rectangular one to mitigate boundary-level distortion and allows the model to access more unique sifting modes. In this paper, we propose a Composite Fusion Attention Transformer (CFAT) that incorporates triangular-rectangular window-based local attention with a channel-based global attention technique in image super-resolution. As a result, CFAT enables attention mechanisms to be activated on more image pixels and captures long-range, multi-scale features to improve SR performance. The extensive experimental results and ablation study demonstrate the effectiveness of CFAT in the SR domain. Our proposed model shows a significant 0.7 dB performance improvement over other state-of-the-art SR architectures. Find the code Here.

Epilepsy is one of the significant neurological disorders affecting nearly 65 million people worldwide. The repeated seizure is characterized as epilepsy. Different algorithms were proposed for efficient seizure detection using intracranial and surface EEG signals. In the last decade, various machine learning techniques based on seizure detection approaches were proposed. This paper discusses different machine learning and deep learning techniques for seizure detection using intracranial and surface EEG signals. A wide range of machine learning techniques such as support vector machine (SVM) classifiers, artificial neural network (ANN) classifier, and deep learning techniques such as a convolutional neural network (CNN) classifier, and long-short term memory (LSTM) network for seizure detection are compared in this paper. The effectiveness of time-domain features, frequency domain features, and time-frequency domain features are discussed along with different machine learning techniques. Along with EEG, other physiological signals such as electrocardiogram are used to enhance seizure detection accuracy which are discussed in this paper. In recent years deep learning techniques based on seizure detection have found good classification accuracy. In this paper, an LSTM deep learning-network-based approach is implemented for seizure detection and compared with state-of-the-art methods. The LSTM based approach achieved 96.5% accuracy in seizure-nonseizure EEG signal classification. Apart from analyzing the physiological signals, sentiment analysis also has potential to detect seizures. Impact Statement- This review paper gives a summary of different research work related to epileptic seizure detection using machine learning and deep learning techniques. Manual seizure detection is time consuming and requires expertise. So the artificial intelligence techniques such as machine learning and deep learning techniques are used for automatic seizure detection. Different physiological signals are used for seizure detection. Different researchers are working on developing automatic seizure detection using EEG, ECG, accelerometer, and sentiment analysis. There is a need for a review paper that can discuss previous techniques and give further research direction. We have discussed different techniques for seizure detection with an accuracy comparison table. It can help the researcher to get an overview of both surface and intracranial EEG-based seizure detection approaches. The new researcher can easily compare different models and decide the model they want to start working on. A deep learning model is discussed to give a practical application of seizure detection. Sentiment analysis is another dimension of seizure detection and summarizing it will give a new prospective to the reader.

Abstract Detection of atrial fibrillation (AF) remains one of the major concerns in the field of medical research. AF is one of the main cause for stroke. AF is characterized by irregular heartbeats and absence of P waves in electrocardiogram (ECG) signal. In this article, we propose a method, combining Poincare plot derived and RR interval‐based features to classify given ECG signal into normal, AF and other rhythms. Classification and regression tree, K‐nearest neighbor, support vector machine, artificial neural network, ResNet18, convolutional neural network (CNN)‐long short term memory (LSTM) are implemented for classification of ECG signal. The class specific accuracies for the three rhythms are computed. Physionet challenge 2017 database is used for evaluation and testing of the developed algorithm. The database has 5154 normal, 771 AF, 2557 other rhythm and 46 noisy signals. Three Poincare derived features viz: SD1, SD2, and ratio of SD1 to SD2, three RR interval features viz: Mean stepping increment of RR interval, approximate entropy and sample entropy are computed and are given to classifiers. During fivefold cross‐validation, CNN‐LSTM classifier showed best result with class specific accuracy for normal of 96.65%, AF of 97.55%, other rhythms of 94.87%, overall accuracy of 96.17% and F1 score of 0.9589. The developed technique can bring change in conventional practice in AF diagnosis and can aid the physician as an assisted tool.

In this paper, we present a novel approach towards customized and automated generation of sports highlights from its extracted events and semantic concepts. A recorded sports video is first divided into slots, based on the game progress and for each slot, an importance-based concept and event-selection is proposed to include those in the highlights. Using our approach, we have successfully extracted highlights from recorded video of cricket match