Human action recognition systems use data collected from a wide range of sensors to accurately identify and interpret human actions. One of the most challenging issues for computer vision is the automatic and precise identification of human activities. A significant increase in feature learning-based representations for action recognition has emerged in recent years, due to the widespread use of deep learning-based features. This study presents an in-depth analysis of human activity recognition that investigates recent developments in computer vision. Augmented reality, human–computer interaction, cybersecurity, home monitoring, and surveillance cameras are all examples of computer vision applications that often go in conjunction with human action detection. We give a taxonomy-based, rigorous study of human activity recognition techniques, discussing the best ways to acquire human action features, derived using RGB and depth data, as well as the latest research on deep learning and hand-crafted techniques. We also explain a generic architecture to recognize human actions in the real world and its current prominent research topic. At long last, we are able to offer some study analysis concepts and proposals for academics. In-depth researchers of human action recognition will find this review an effective tool.

Feature selection is an important topic in machine learning. In recent years, via integrating ensemble learning, the ensemble learning based feature selection approach has been proposed and studied. The general idea is to generate multiple diverse feature selectors and combine their outputs. This approach is superior to conventional feature selection methods in many aspects. Among them, its most prominent advantage is the ability to handle stability issue that is usually poor in existing feature selection methods. This review covers different issues related to ensemble learning based feature selection, which include the main modules, the stability measurement, etc. To the best of our knowledge, this is the first review that focuses on ensemble feature selection. It can be a useful reference in the literature of feature selection.

Background: No reported study has compared clinical and radiologic outcomes between an all-inside arthroscopic modified Broström operation (MBO) and an open MBO. The purpose of this study was to compare clinical and radiologic outcomes of all-inside arthroscopic and open MBOs. Methods: From August 2012 to July 2014, 48 patients were included. They were divided into 2 groups: all-inside arthroscopic MBO (25 patients) and open MBO (23 patients). The American Orthopaedic Foot & Ankle Society (AOFAS) ankle–hindfoot score, visual analog scale (VAS) score, and Karlsson score were used to evaluate clinical outcomes. Anterior talar translation and talar tilt were used to evaluate radiologic outcomes. All patients had lateral ankle instability. MBO was performed in 87 patients. Of these, 50 patients met the inclusion criteria. All patients had giving way, persistent pain, and an inability to resume their preinjury activity level for more than 6 months. Patients were randomized into 2 groups, all-inside arthroscopic MBO and open MBO, using a permuted block randomization method. Clinical outcome evaluations were performed preoperatively, at 6 weeks and 6 months postoperatively, and at a final follow-up at a minimum of 12 months postoperatively using the Karlsson score, the AOFAS ankle-hindfoot score, and pain VAS scores. Radiologic outcome evaluations were performed preoperatively and at 1 year postoperatively at final follow-up using anterior talar translation, and talar tilt angle. Results: After randomization, 25 ankles were allocated to the all-inside arthroscopic MBO group and 25 to the open MBO group. Two ankles in the open MBO group were excluded from the analysis because they were lost to follow-up. Thus, evaluations were performed for 25 ankles in the all-inside arthroscopic MBO group and 23 in the open MBO group. There was no difference in age, gender, symptom duration, preoperative AOFAS, VAS, Karlsson scores, anterior talar translation, or talar tilt between the 2 groups (all P > .05). At the final follow-up, the AOFAS, VAS, and the Karlsson scores had improved significantly in both groups ( P < .001). There was no difference in the Karlsson, AOFAS, or VAS scores, anterior talar translation, or talar tilt between the 2 groups at final follow-up (all P > .05). Conclusions: There was no difference in the clinical or radiologic outcome between the all-inside arthroscopic MBO and open MBO for the treatment of lateral ankle instability at up to 1 year after surgery. An all-inside arthroscopic MBO should be considered carefully in patients who have lateral ankle instability. Level of Evidence: Level I, randomized controlled trial.

Depression is a serious psychiatric disorder that restricts an individual's ability to work properly in both their daily and professional lives. Usually, the diagnosis of depression often needs a thorough assessment by an expert professional. Recently, significant consideration has been given to automatic depression prediction for more reliable and efficient depression investigation. In this paper, we propose a novel framework to estimate the depression level from video data by employing a two-stream deep spatiotemporal network. Our approach extracts spatial information using the Inception-ResNet-v2 network. In contrast, we introduce a volume local directional number (VLDN) based dynamic feature descriptor to capture facial motions. Then, the feature map obtained from the VLDN is fed into a convolutional neural network (CNN) to obtain more discriminative features. Additionally, we designed a multilayer bidirectional long short-term memory (Bi-LSTM) model to obtain temporal information by integrating the temporal median pooling (TMP) approach into the model. The TMP approach is employed on the temporal fragments of spatial and temporal features. Finally, extensive experimental analysis of two challenging datasets, AVEC2013 and AVEC2014, demonstrates that the proposed approach shows promising performance compared to the existing approaches for depression level prediction.

Recent developments in smartphones have increased the processing capabilities and equipped these devices with a number of built-in multimodal sensors, including accelerometers, gyroscopes, GPS interfaces, Wi-Fi access, and proximity sensors. Despite the fact that numerous studies have investigated the development of user-context aware applications using smartphones, these applications are currently only able to recognize simple contexts using a single type of sensor. Therefore, in this work, we introduce a comprehensive approach for context aware applications that utilizes the multimodal sensors in smartphones. The proposed system is not only able to recognize different kinds of contexts with high accuracy, but it is also able to optimize the power consumption since power-hungry sensors can be activated or deactivated at appropriate times. Additionally, the system is able to recognize activities wherever the smartphone is on a human’s body, even when the user is using the phone to make a phone call, manipulate applications, play games, or listen to music. Furthermore, we also present a novel feature selection algorithm for the accelerometer classification module. The proposed feature selection algorithm helps select good features and eliminates bad features, thereby improving the overall accuracy of the accelerometer classifier. Experimental results show that the proposed system can classify eight activities with an accuracy of 92.43%.